[{"id":"cloudflare-workers","provider":"cloudflare","name":"Cloudflare Workers","category":"serverless","also":["edge-compute"],"tagline":"Run your app's code close to your users, with no servers to manage.","description":"Cloudflare Workers runs your code on Cloudflare's worldwide network, so you never pick a region or look after a server. It works with JavaScript, TypeScript, Python and WASM (a format that lets other languages run on the web). If you're building an API for your app or a scheduled job that sends a weekly email, a Worker can handle it. Built-in connections let it reach Cloudflare's storage, databases and queues directly. Pick it when you want fast responses for users anywhere without setting up hosting. There's a free tier, then $5 a month plus usage.","useCases":["Run API backends for your app","Host full-stack apps close to users","Run scheduled jobs, like weekly emails"],"pricing":"Free tier, then Workers Paid from $5/month plus usage.","status":"ga","url":"https://developers.cloudflare.com/workers/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-pages","provider":"cloudflare","name":"Cloudflare Pages","category":"hosting","tagline":"Publish your website from Git, with a preview link for every change.","description":"Cloudflare Pages hosts your website straight from a Git repository, the place where your code is saved and tracked. If you've built a landing page or marketing site, push a change and it goes live, with a preview link so you can check updates first. It can also run small pieces of backend code through Pages Functions, which are powered by Workers. Cloudflare now steers new full-stack projects toward Workers with static assets, so Pages is best for simpler sites. There's a free tier.","useCases":["Publish landing pages and marketing sites","Preview changes with a link before going live","Host simple sites from a Git repository"],"pricing":"Free tier; paid usage follows Workers pricing.","status":"ga","url":"https://developers.cloudflare.com/pages/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-workers-builds","provider":"cloudflare","name":"Workers Builds","category":"devtools","also":["hosting"],"tagline":"Automatically build and publish your Worker every time you push code.","description":"Workers Builds connects a GitHub or GitLab repository to a Worker, so each time you push code it is built and published for you. If your project lives on GitHub, your live app updates without you running any commands. Other branches get preview builds, so you can test changes before they go live. It also handles monorepos, which are single repositories holding several projects. It's included with Workers plans, and build minutes have limits that depend on your plan.","useCases":["Publish your app automatically on each code push","Test changes on preview builds first","Deploy several projects from one repository"],"pricing":"Included with Workers plans; build minutes have plan limits.","status":"ga","url":"https://developers.cloudflare.com/workers/ci-cd/builds/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-workers-for-platforms","provider":"cloudflare","name":"Workers for Platforms","category":"serverless","also":["edge-compute"],"tagline":"Let your customers run their own code inside your product.","description":"Workers for Platforms lets your customers deploy their own code on top of your product, each in a separate, isolated space with its own limits. If you're building a site builder or a SaaS tool where customers write custom scripts, this is how you run that code safely. A dispatch Worker, which works like a traffic director, sends each request to the right customer's code. Pick it over plain Workers when you're hosting code from many customers, not just your own. It's a paid add-on with a base fee plus usage, and you contact sales for large scale.","useCases":["Let customers run their own custom scripts","Build a site builder that runs customer code","Run separate functions for each customer"],"pricing":"Paid add-on plan; base fee plus usage. Contact sales for scale.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-dynamic-workers","provider":"cloudflare","name":"Dynamic Workers","category":"serverless","also":["agents"],"tagline":"Create throwaway, isolated spaces to safely run code you did not write.","description":"Dynamic Workers creates sandboxed Workers, which are small locked-down spaces for code, on demand. The code can come from your app or from an AI agent while it's running. If your product lets an AI write code or lets users add plugins, you can run that code without putting the rest of your app at risk. Pick it over regular Workers when the code is generated or untrusted rather than written ahead of time. It also pairs with Containers to form Cloudflare Sandboxes. It's included with Workers Paid.","useCases":["Run AI-generated code without risking your app","Safely run plugins that users add","Create a fresh sandbox for each request"],"pricing":"Included with Workers Paid; see docs for limits.","status":"ga","url":"https://developers.cloudflare.com/dynamic-workers/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-sandboxes","provider":"cloudflare","name":"Cloudflare Sandboxes","category":"agents","also":["containers"],"tagline":"A safe computer for your AI agent to run code, commands and files.","description":"Cloudflare Sandboxes gives an AI agent or app a secure place to run code, shell commands (typed instructions to a computer) and work with files. If you're building a coding assistant or a tool that analyzes uploaded data, the agent can try things out without touching your real systems. It's built on Containers and Dynamic Workers, which makes it a good fit when an agent needs a full Linux environment. You pay by usage, billed through those two products.","useCases":["Give a coding assistant a safe place to run code","Let an agent analyze uploaded data files","Offer interactive dev environments to users"],"pricing":"Usage-based through Containers and Workers.","status":"ga","url":"https://developers.cloudflare.com/sandbox/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-containers","provider":"cloudflare","name":"Cloudflare Containers","category":"containers","also":["serverless"],"tagline":"Run heavier software packaged as containers, launched from Workers.","description":"Cloudflare Containers runs container images, which are software packaged with everything it needs to run, on Cloudflare's network. You start and control them from Workers. If your app needs to convert video with FFmpeg or run a machine learning task, that kind of job may call for a full Linux environment, and a container gives you one. Instances start on demand and you're billed only for active time. It needs Workers Paid, plus usage charges for CPU, memory and disk time.","useCases":["Run video conversion with FFmpeg","Run heavy machine learning jobs in a full environment","Start a separate container for each user"],"pricing":"Workers Paid plus usage for CPU, memory and disk time.","status":"ga","url":"https://developers.cloudflare.com/containers/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-durable-objects","provider":"cloudflare","name":"Durable Objects","category":"serverless","also":["nosql","realtime"],"tagline":"Small programs with their own storage, one per chat room, user or game.","description":"Durable Objects are small programs that each keep their own data and live in one place, found by a unique ID. If you're building a chat room, a shared document or a rate limiter (a counter that stops people sending too many requests), each one gets its own object. Because only one copy exists, its storage stays consistent, so two people never see conflicting numbers. Pick them over a regular database when users need live coordination. There's a free tier for the version backed by SQLite, a lightweight database. Workers Paid bills on requests and duration.","useCases":["Power live collaboration in shared documents","Limit how often each user can send requests","Keep separate saved data for each user"],"pricing":"Free tier (SQLite-backed); Workers Paid with request and duration billing.","status":"ga","url":"https://developers.cloudflare.com/durable-objects/","source":"https://developers.cloudflare.com/products/","gem":"Tiny stateful servers per chat room, document or game, spun up on demand."},{"id":"cloudflare-queues","provider":"cloudflare","name":"Cloudflare Queues","category":"queues","tagline":"Hand off slow jobs to run later, with no fee for moving data out.","description":"Cloudflare Queues lets one part of your app drop a message into a line so another part can handle it later. If a signup should trigger a welcome email, or you receive webhooks (automatic notifications from other services) in bursts, a queue keeps your app quick. It supports batching, retries, delays and dead-letter queues, a holding spot for messages that keep failing. Workers can read from it, and so can anything over HTTP. It charges no egress fees, meaning no charge for data leaving. Pricing is per operation, with Free and Paid plans.","useCases":["Send welcome emails after signups in the background","Handle bursts of incoming webhooks smoothly","Hold failed jobs for retries or review"],"pricing":"Usage-based per operation; Free and Paid plans.","status":"ga","url":"https://developers.cloudflare.com/queues/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-workflows","provider":"cloudflare","name":"Cloudflare Workflows","category":"workflows","also":["serverless"],"tagline":"Run multi-step processes that survive crashes and can wait for days.","description":"Cloudflare Workflows lets you write a long process as a series of steps in code. Progress is saved after each step, so the process can pick up where it left off after a restart. If you're building an order process or an approval flow where a person must say yes, a step can sleep for days or wait for an outside event. Failed steps can retry. Pick it over a simple queue when the order of steps and saved progress matter. There's a free tier, and Workers Paid bills for CPU time and step storage.","useCases":["Run order processes that survive restarts","Pause a process for days awaiting approval","Retry failed steps in a multi-step process"],"pricing":"Free tier; Workers Paid with CPU time and step storage billing.","status":"ga","url":"https://developers.cloudflare.com/workflows/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-browser-run","provider":"cloudflare","name":"Browser Run","category":"agents","also":["serverless"],"tagline":"Control a web browser from code to take screenshots, make PDFs and scrape.","description":"Browser Run, formerly Browser Rendering, lets your code control headless Chrome, which is a web browser with no visible window. If your app needs to turn a page into a PDF or screenshot, Quick Actions does it with one HTTP call, and it also covers markdown and scraping. For bigger jobs like testing or agents browsing the web, Browser Sessions work with tools such as Puppeteer and Playwright. Pick it over running your own browser when you don't want to maintain one. There's a free tier, then you pay for browser time used.","useCases":["Turn web pages into PDFs or screenshots","Scrape information from websites","Let an AI agent browse the web"],"pricing":"Free tier, then pay for browser time used.","status":"ga","url":"https://developers.cloudflare.com/browser-run/","source":"https://developers.cloudflare.com/products/","gem":"A real headless Chrome you can drive from a few lines of code."},{"id":"cloudflare-workers-vpc","provider":"cloudflare","name":"Workers VPC","category":"networking","also":["serverless"],"tagline":"Let your Workers securely reach private services in other clouds.","description":"Workers VPC lets your Workers talk to private services that aren't open to the public internet, such as ones in another cloud or on your own servers. If your app needs to call an internal API or a private MCP server (a tool connection for AI agents), you link it through a Tunnel, a secure connection out from your network. VPC Services connect to one specific host and port, while VPC Networks give broader access. It supports HTTP and raw TCP. It's currently in beta and free on all Workers plans during the beta.","useCases":["Call private internal APIs from your Workers","Connect AI tools to private MCP servers","Link apps running across different clouds"],"pricing":"Free during beta on all Workers plans.","status":"beta","url":"https://developers.cloudflare.com/workers-vpc/","source":"https://developers.cloudflare.com/workers-vpc/"},{"id":"cloudflare-email-sending","provider":"cloudflare","name":"Email Sending","category":"email","tagline":"Send your app's receipts, alerts and account emails to anyone.","description":"Email Sending is part of Cloudflare Email Service, and it lets your app send automated emails like receipts, order confirmations and alerts. If your store needs to email a customer right after they buy, you can trigger it from a Cloudflare Worker (a small piece of code that runs on Cloudflare), a REST API (a standard web request) or SMTP (the classic email-sending method). It also gives you deliverability guidance and analytics, so you can see how your messages are doing. It is in beta, and sending to any recipient requires the Workers Paid plan, so check the docs for current rates.","useCases":["Send receipts and order confirmations automatically","Email alerts from your app to users","Check delivery stats for your app emails"],"pricing":"Sending to any recipient needs Workers Paid; see docs for rates.","status":"beta","url":"https://developers.cloudflare.com/email-service/","source":"https://developers.cloudflare.com/email-service/"},{"id":"cloudflare-email-routing","provider":"cloudflare","name":"Email Routing","category":"email","tagline":"Make custom addresses like help@yourdomain and forward or process the mail.","description":"Email Routing lets you create addresses on your own domain and decide what happens to the mail sent to them. If you want help@yourcompany.com to land in your personal inbox, you can forward it in a few clicks. If you want something smarter, like turning each incoming message into a support ticket, you can hand it to an Email Worker, which is a small piece of code that reads the message. It is free on all plans, which makes it an easy pick over paid mailbox forwarding tools. The docs now list it under Email Service.","useCases":["Create help@ addresses on your own domain","Forward incoming mail to your inbox","Turn incoming emails into support tickets"],"pricing":"Free on all plans.","status":"ga","url":"https://developers.cloudflare.com/email-service/","source":"https://developers.cloudflare.com/email-service/","gem":"Free custom email addresses that forward anywhere, or run code on every message."},{"id":"cloudflare-flagship","provider":"cloudflare","name":"Flagship","category":"feature-flags","also":["devtools"],"tagline":"Turn app features on or off for chosen users, without redeploying.","description":"Flagship is a feature flag tool, meaning it lets you switch parts of your app on or off without publishing new code. If you built a new checkout page and want only 10 percent of visitors to see it first, you set a percentage rollout and widen it as it proves stable. Flags can be simple on/off values, text, numbers or JSON, with rules for who gets what. It runs close to your users on Cloudflare Workers, and also works with OpenFeature, an open standard, for TypeScript, Python and Go. It is in beta, and the docs do not state pricing.","useCases":["Roll out a new checkout to some visitors","Run A/B tests with different variants","Turn off a broken feature quickly"],"pricing":"Not stated in docs.","status":"beta","url":"https://developers.cloudflare.com/flagship/","source":"https://developers.cloudflare.com/flagship/"},{"id":"cloudflare-workers-ai","provider":"cloudflare","name":"Workers AI","category":"ai-models","also":["serverless"],"tagline":"Run AI models for chat, images and speech, with no servers to manage.","description":"Workers AI lets you use open AI models for chat, embeddings (numbers that capture meaning), images, speech and vision, without renting or managing graphics chips yourself. If you want a chatbot in your app or a way to turn voice notes into text, you call a model from a Cloudflare Worker or a regular web API. You get a free daily allowance, then pay for what you use. Pick it when you want AI running on the same platform as the rest of your app, billed only as it is used.","useCases":["Add a chatbot to your app","Turn voice notes into text","Generate images for your product"],"pricing":"Free daily allowance, then pay per neuron or token.","status":"ga","url":"https://developers.cloudflare.com/workers-ai/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-ai-gateway","provider":"cloudflare","name":"AI Gateway","category":"ai-gateway","tagline":"One front door for your AI calls, with logs, caching and spending visibility.","description":"AI Gateway sits between your app and the AI providers you call, so you can see and control what is happening. If your chatbot is getting expensive, it can show you what each request costs, save repeated answers so you do not pay twice, and cap how often requests go out. It can also retry failures or switch to a backup provider if one goes down. It works with Workers AI and third-party providers. Core features are free on all plans, though storing logs has limits.","useCases":["Track what each AI request costs","Reuse saved AI answers to avoid paying twice","Switch to a backup AI provider when one fails"],"pricing":"Core features free on all plans; log storage has limits.","status":"ga","url":"https://developers.cloudflare.com/ai-gateway/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-mcp-server-portals","provider":"cloudflare","name":"MCP Server Portals","category":"agents","also":["zero-trust"],"tagline":"Put all your MCP tool servers behind one address with access rules.","description":"MCP Server Portals gather your MCP servers, which are the connectors that let AI assistants use outside tools, behind one web address. If your team has several of them and you worry about who can reach what, you can apply Cloudflare Access rules (sign-in and permission checks) to all of them in one place. It also logs the traffic, so you can review what the AI tools did. The docs list it under AI controls, and it is included with your Cloudflare One plan rather than sold separately.","useCases":["Put all MCP tool servers behind one address","Set who can reach each AI tool","Review logs of what AI tools did"],"pricing":"Part of Cloudflare One; included with plan.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/access-controls/ai-controls/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-vectorize","provider":"cloudflare","name":"Vectorize","category":"vector","also":["ai-search"],"tagline":"Store and search AI embeddings so your app can find things by meaning.","description":"Vectorize is a database for embeddings, which are lists of numbers that represent the meaning of text or images. If you want search that understands \"cheap flights\" and \"low-cost airfare\" as the same idea, you store embeddings here and look up the closest matches. It is also how you feed relevant documents to an AI for RAG, where the AI answers using your own content. You can filter by extra details, and it pairs well with Workers AI. There is a free tier, and Workers Paid bills by stored and queried vector dimensions.","useCases":["Search by meaning, like cheap flights","Feed your documents to an AI for answers","Recommend similar items to users"],"pricing":"Free tier; Workers Paid billed on stored and queried vector dimensions.","status":"ga","url":"https://developers.cloudflare.com/vectorize/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-ai-search","provider":"cloudflare","name":"AI Search","category":"ai-search","also":["vector"],"tagline":"Let people ask questions about your content, with no setup pipeline.","description":"AI Search, formerly called AutoRAG, lets visitors or AI agents ask plain-language questions about your own files and get answers drawn from them. If you have a help center and want a search box that actually understands questions, you point it at your content and it keeps its index up to date automatically. You can query it from a Worker, a REST API or a built-in MCP endpoint. It also supports hybrid search, filtering, image search and OCR (reading text from images). It is available on all plans, though pricing is not on the product page.","useCases":["Let visitors ask questions about your docs","Give AI agents a knowledge base to query","Search separate file sets for each customer"],"pricing":"Available on all plans; pricing not on product page.","status":"ga","url":"https://developers.cloudflare.com/ai-search/","source":"https://developers.cloudflare.com/ai-search/"},{"id":"cloudflare-agent-tracing","provider":"cloudflare","name":"Agent tracing","category":"observability","also":["agents"],"tagline":"See every step your AI agent took, from model calls to tool runs.","description":"Agent tracing shows you what your AI agent did, step by step. If an agent is slow or giving odd answers, you can replay a session and view a timeline of each model call, tool run and approval, with token counts (the units AI usage is billed in). It works with Think, Flue, AI SDK or custom OpenTelemetry setups, and traces can be sent to any OTLP destination, a common format for monitoring tools. It is part of Cloudflare Agents observability, launched August 2026, and uses Workers observability billing.","useCases":["Replay an agent session step by step","Find out why an agent is slow","Track token use to review AI costs"],"pricing":"Uses Workers observability billing.","status":"ga","url":"https://developers.cloudflare.com/agents/runtime/operations/observability/tracing/","source":"https://blog.cloudflare.com/agents-on-cloudflare/"},{"id":"cloudflare-agents-sdk","provider":"cloudflare","name":"Cloudflare Agents SDK","category":"agents","also":["serverless"],"tagline":"Build AI agents that remember things and keep running on Cloudflare.","description":"The Cloudflare Agents SDK is a toolkit for building AI agents that keep their own memory between conversations. If you want a chat assistant that remembers each user and can also wake up on a schedule to do tasks, this gives you that structure. Each agent runs as a Durable Object, a small piece of Cloudflare code that holds its own saved data. It supports sessions, live WebSocket connections, scheduling and observability, and can host MCP servers. Pick it when you want agents and their data in one place. Billing is usage-based through Workers and Durable Objects.","useCases":["Build chat agents that remember each user","Schedule agents to run tasks on their own","Host MCP servers alongside your agents"],"pricing":"Usage-based through Workers and Durable Objects.","status":"ga","url":"https://developers.cloudflare.com/agents/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-agent-memory","provider":"cloudflare","name":"Agent Memory","category":"agents","also":["ai-platform"],"tagline":"Give your AI agent memory that picks up what matters from conversations.","description":"Agent Memory is a long-term memory for AI agents and apps. If your assistant should remember that a customer prefers short answers or uses a certain product, it can pick up that knowledge from conversations, sort it, and bring it back later. That makes personalization possible without you building the memory system yourself. The product index describes it this way, though details beyond that line were not verified. It is in beta, and pricing is not stated, so check Cloudflare before planning around it.","useCases":["Let an assistant recall past customer preferences","Personalize answers based on earlier chats","Save useful facts from past conversations"],"pricing":"Not stated.","status":"beta","url":"https://developers.cloudflare.com/agent-memory/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-web-search-api","provider":"cloudflare","name":"Web Search API","category":"agents","also":["ai-search"],"tagline":"Give your AI agent live web results so its answers stay current.","description":"Web Search API returns live results from the web for AI agents and apps. If your assistant needs to answer \"what changed in this law last week?\", it can look it up instead of relying only on what the AI learned in training. This is called grounding, and it helps answers stay fresh and checkable. It can also add current information to RAG, where an AI answers using retrieved documents. The product index lists it, but detail pages were not verified. It is in beta, and pricing is not stated.","useCases":["Answer questions with live web results","Check recent changes before answering","Ground agent answers in current web pages"],"pricing":"Not stated.","status":"beta","url":"https://developers.cloudflare.com/web-search/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-agent-lee","provider":"cloudflare","name":"Agent Lee","category":"ai-apps","also":["management"],"tagline":"An AI helper in the Cloudflare dashboard that you can just talk to.","description":"Agent Lee is an AI assistant built into the Cloudflare dashboard, the web page where you manage your Cloudflare account. You type what you want in plain English, and it helps you find the right screen, set up services or take actions for you. If you are setting up Cloudflare for the first time and are not sure where a setting lives, you can ask instead of digging through menus. Pick it when you want guided help with configuration without reading all the docs first. It is included with the dashboard.","useCases":["Ask where a setting lives in plain English","Get help setting up services","Have actions taken for you in the dashboard"],"pricing":"Included with the dashboard.","status":"ga","url":"https://developers.cloudflare.com/agent-lee/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloudflare-wallets","provider":"cloudflare","name":"Cloudflare Wallets","category":"specialized","also":["agents"],"tagline":"Claim a readable wallet name so your AI agents can make payments.","description":"Cloudflare Wallets gives your account and its AI agents a stable, human-readable name for payments, so nobody has to deal with long strings of characters. If you are building an agent that needs to pay for things on its own, such as access to a resource that charges through x402, a way for websites to ask for payment over the web, a wallet handle gives it a consistent identity. It is a good fit when agent payments are part of your product plan. It is currently in beta, and pricing is not stated.","useCases":["Give each AI agent a readable payment name","Let agents pay for paid online resources","Let websites ask agents for payment"],"pricing":"Not stated.","status":"beta","url":"https://developers.cloudflare.com/wallets/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-r2","provider":"cloudflare","name":"Cloudflare R2","category":"object-storage","tagline":"Store your app's files and images, with no fee to download them.","description":"Cloudflare R2 is online storage for files like images, videos, backups and website assets. If your app lets people upload photos or videos, R2 is where those files can live. It works with the S3 API, the common language many tools use to talk to file storage, so tools built for Amazon S3 can usually talk to it. Pick it over the obvious alternative when download costs worry you, since R2 has no egress fees, meaning no charge when files leave storage. It also includes tools to move existing files over, either in bulk or gradually.","useCases":["Store photos and videos users upload","Host website images and downloads","Keep nightly backups of your data"],"pricing":"Free tier, then per GB-month and operations; no egress fees.","status":"ga","url":"https://developers.cloudflare.com/r2/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-basin-catalog","provider":"cloudflare","name":"Basin Catalog","category":"data-pipelines","also":["warehouse","object-storage"],"tagline":"Keep track of analytics tables stored in R2 using the open Iceberg format.","description":"Basin Catalog is a managed directory for data tables kept in Cloudflare R2, using Apache Iceberg, an open format for large analytics tables. You create and manage tables here, then query them with compatible tools. If you are collecting lots of product data and want to analyze it without locking it into one vendor's warehouse, this gives you a tidy home for it. Pick it when you want open-format tables sitting in your own storage. You pay the usual R2 storage and operation charges.","useCases":["Track analytics tables in your storage","Keep product data in an open format","Query tables with other compatible tools"],"pricing":"R2 storage and operation charges; see docs.","status":"ga","url":"https://developers.cloudflare.com/basin/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-basin-pipelines","provider":"cloudflare","name":"Basin Pipelines","category":"data-pipelines","also":["queues"],"tagline":"Collect app events, clean them up with SQL, and save them to R2.","description":"Basin Pipelines, formerly Cloudflare Pipelines, takes a steady flow of events from your app and stores them as organized data. Say you want to record every click or signup on your site: your app sends each event over HTTP or from a Worker, a small program running on Cloudflare. SQL, a common language for working with data, checks and enriches the events. Then they are written to R2 as Iceberg tables, an open analytics format, or as Parquet (a compact data file format) and JSON files. It requires the Workers Paid plan.","useCases":["Record every click and signup in your app","Clean up event data with SQL","Save events as organized files in storage"],"pricing":"Requires Workers Paid; rates on pricing page.","status":"ga","url":"https://developers.cloudflare.com/pipelines/","source":"https://developers.cloudflare.com/pipelines/"},{"id":"cloudflare-basin-sql","provider":"cloudflare","name":"Basin SQL","category":"warehouse","also":["data-pipelines"],"tagline":"Ask questions of your stored data with SQL, no servers to run.","description":"Basin SQL lets you run SQL, the standard language for asking questions of data, over tables stored in R2. There is no cluster, meaning a group of servers, for you to set up or look after. If you have been saving product events and want to know how many people signed up last week, you can query them directly. It is built for tables in Basin Catalog, so it fits best when your data already lives there. Pricing is not stated in the catalog index, so check the docs.","useCases":["Ask questions of stored product data","Count signups from last week","Check pipeline output with SQL queries"],"pricing":"Not stated on index; see docs.","status":"ga","url":"https://developers.cloudflare.com/basin/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-k2-streams","provider":"cloudflare","name":"K2 Streams","category":"queues","also":["data-pipelines"],"tagline":"A replayable log of events that many parts of your app can read.","description":"K2 Streams is a running record of events that one part of your app writes and other parts read, each at its own pace. If your store needs to tell billing, email and analytics about every new order, you can write the order once and let each one pick it up. Records are kept for up to 30 days, so a reader that falls behind can catch up later. Your code writes through a Workers connection or plain HTTP. It is in public beta with a 10 GB cap per account, and rates are not published yet.","useCases":["Send each new order to several services","Let billing and email read the same events","Replay events for readers that fell behind"],"pricing":"Workers Paid; rates not published during beta.","status":"beta","url":"https://developers.cloudflare.com/k2/","source":"https://developers.cloudflare.com/k2/"},{"id":"cloudflare-hyperdrive","provider":"cloudflare","name":"Hyperdrive","category":"sql","also":["networking"],"tagline":"Make your existing Postgres or MySQL database fast from Cloudflare Workers.","description":"Hyperdrive makes an existing database respond faster when your code runs on Cloudflare Workers, small programs that run close to your users. It works with Postgres and MySQL, two popular databases, hosted on any cloud. If your app's data already lives in a Postgres database elsewhere, Hyperdrive reuses connections and caches common read results so pages load faster. Pick it when you want to keep your current database instead of moving everything. It is included with Workers plans.","useCases":["Speed up Postgres from Workers","Connect Workers to an existing database","Cache common reads for faster pages"],"pricing":"Included with Workers plans.","status":"ga","url":"https://developers.cloudflare.com/hyperdrive/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-workers-kv","provider":"cloudflare","name":"Workers KV","category":"kv-cache","tagline":"A fast lookup table for small bits of data, available worldwide.","description":"Workers KV is a key-value store, which works like a giant list where you save a value under a name and fetch it back by that name. Use it for things like feature settings, login session tokens or cached lookups that your app reads far more often than it changes. Reads are fast from code running on Cloudflare around the world. The tradeoff is that it is eventually consistent, so an update can take a moment to show up everywhere. Pick it for read-heavy data, not for things that must be exact instantly.","useCases":["Store feature settings and config values","Keep login session tokens handy","Serve cached lookups fast worldwide"],"pricing":"Free tier, then per read, write and GB stored.","status":"ga","url":"https://developers.cloudflare.com/kv/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-d1","provider":"cloudflare","name":"Cloudflare D1","category":"sql","tagline":"A ready-to-use SQL database for your app, built on SQLite.","description":"Cloudflare D1 is a managed database built on SQLite, a lightweight and widely used database, that your Cloudflare Workers code can talk to directly. If you are prototyping an app and need a place to store users and orders, you can start on D1's free tier. It is also designed for many small databases, such as one for each customer. Time Travel lets you restore a database to an earlier moment, and read replication keeps extra copies for faster reads. Pick it when you want a simple SQL database without running a server.","useCases":["Store users and orders in a SQL database","Give each customer their own database","Prototype an app without running servers"],"pricing":"Free tier, then per rows read, written and storage.","status":"ga","url":"https://developers.cloudflare.com/d1/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-artifacts","provider":"cloudflare","name":"Artifacts","category":"devtools","also":["object-storage","agents"],"tagline":"Versioned file storage that speaks Git, made for apps and AI agents.","description":"Artifacts is storage for files that keeps a history of every change, using Git, the version tracking system most developers already use. If your AI agent writes code or documents for each user or task, you can give each one its own repository, which is a project folder with history. You create them from Workers, a web API or ordinary Git tools, and you can import existing repos or connect them to Workers Builds, Cloudflare's tool for building and deploying code. It is listed as beta, and the docs do not state pricing.","useCases":["Give each AI agent its own repository","Save agent-written code and documents","Track every change to files"],"pricing":"Not stated in docs.","status":"beta","url":"https://developers.cloudflare.com/artifacts/","source":"https://developers.cloudflare.com/artifacts/"},{"id":"cloudflare-analytics-engine","provider":"cloudflare","name":"Workers Analytics Engine","category":"analytics","also":["observability"],"tagline":"Track custom numbers from your app and query them with SQL.","description":"Workers Analytics Engine lets your Cloudflare Workers code record custom data points, then ask questions about them with SQL, the standard data-query language. If you want to meter how much each customer uses your product, or track your own app metrics, you send a data point each time something happens. It supports unlimited cardinality, meaning you can break results down by as many different values, like user IDs, as you need. Many people pair it with Grafana, a dashboard tool, to chart the results. It has a free tier, then charges per data points written and per query.","useCases":["Meter how much each customer uses","Track your own app metrics","Chart results in a dashboard tool"],"pricing":"Free tier, then per data points written and queries.","status":"ga","url":"https://developers.cloudflare.com/analytics/analytics-engine/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-secrets-store","provider":"cloudflare","name":"Secrets Store","category":"secrets","tagline":"Keep passwords and API keys in one safe place and share them across apps.","description":"Secrets Store is a locked vault for your passwords and API keys, the private codes that let your app talk to other services. If you run several Cloudflare Workers, which are small programs that run on Cloudflare's network, you can save a key once and let all of them use it. Instead of pasting the same key into every Worker, you keep one copy. When you need to change a key, you update it in one place. It is included with your account.","useCases":["Keep API keys in one safe place","Share one key across several apps","Update a key in one spot"],"pricing":"Included with account.","status":"ga","url":"https://developers.cloudflare.com/secrets-store/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-images","provider":"cloudflare","name":"Cloudflare Images","category":"images","also":["cdn"],"tagline":"Store images and resize them automatically for any screen.","description":"Cloudflare Images stores your pictures and delivers them fast, resizing and converting them on demand. If your app lets people upload photos, you can keep them here and serve a small version for phones and a larger one for desktops, without making each copy yourself. It can also resize and convert images that live somewhere else. Pricing is based on how many images you store and deliver, and transformations have a free allowance.","useCases":["Store user-uploaded photos","Serve right-sized images for each screen","Convert images to other formats"],"pricing":"Per image stored and delivered; transformations have a free allowance.","status":"ga","url":"https://developers.cloudflare.com/images/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-stream","provider":"cloudflare","name":"Cloudflare Stream","category":"video","tagline":"Upload video once and play it smoothly anywhere, on demand or live.","description":"Cloudflare Stream takes care of storing, converting and playing video so you do not have to build that yourself. If you sell an online course or want short clips inside your app, you upload the video and Stream prepares it to play well on any device and connection speed. It also handles live broadcasts and gives you viewing analytics. You pay per minute of video stored and per minute delivered, which makes costs easy to estimate.","useCases":["Sell and host online course videos","Show short clips inside your app","Broadcast live video to viewers"],"pricing":"Per minute stored and per minute delivered.","status":"ga","url":"https://developers.cloudflare.com/stream/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-realtimekit","provider":"cloudflare","name":"RealtimeKit","category":"realtime","also":["video"],"tagline":"Add video and voice calls to your web or mobile app with ready-made parts.","description":"RealtimeKit is a toolkit for adding audio and video calls to your app, with ready-made screens so you do not design them from scratch. If you are building a coaching app, a webinar tool or a voice chat room, you can drop in a meeting experience instead of building one. It can also record calls and write transcripts if you want. It runs on Cloudflare's own call network and was formerly part of Cloudflare Calls. Pricing depends on usage.","useCases":["Add video calls to your app","Host webinars with ready-made screens","Build voice chat rooms"],"pricing":"Usage-based; see Realtime pricing.","status":"ga","url":"https://developers.cloudflare.com/realtime/realtimekit/","source":"https://developers.cloudflare.com/realtime/"},{"id":"cloudflare-realtime-sfu","provider":"cloudflare","name":"Realtime SFU","category":"realtime","also":["video"],"tagline":"Build your own video call or live stream app with full control.","description":"Realtime SFU is the behind-the-scenes traffic director for video and audio calls. An SFU, short for selective forwarding unit, receives each person's stream and passes it only to the people who should get it. If you are building a custom conferencing tool or a multiplayer app with live media, you decide who can send and receive, and your app handles sign-in and permissions. Pick it over RealtimeKit when you want control rather than ready-made screens. It was previously called Cloudflare Calls, and billing is based on data sent out, with a free allowance.","useCases":["Build a custom video conferencing tool","Stream live video with your own rules","Power multiplayer apps with live media"],"pricing":"Usage-based on egress; free allowance included.","status":"ga","url":"https://developers.cloudflare.com/realtime/sfu/","source":"https://developers.cloudflare.com/realtime/"},{"id":"cloudflare-turn-service","provider":"cloudflare","name":"Cloudflare TURN","category":"realtime","also":["networking"],"tagline":"A backup route that keeps video calls connected through firewalls.","description":"Cloudflare TURN is a relay that passes call traffic along when a direct connection between two people is blocked. Some office networks and home routers, the boxes that connect devices to the internet, refuse direct links, which can leave a video call stuck on a blank screen. If your app has calls, this relay keeps them working for those users. You do not need to buy it separately if you use RealtimeKit or the Realtime SFU, since it is included with both. Usage is billed on data sent out, with a free allowance.","useCases":["Keep calls working behind strict firewalls","Relay calls when direct links fail","Rescue video calls stuck on blank screens"],"pricing":"Usage-based on egress; free allowance included.","status":"ga","url":"https://developers.cloudflare.com/realtime/turn/","source":"https://developers.cloudflare.com/realtime/"},{"id":"cloudflare-moq-relay","provider":"cloudflare","name":"MoQ Relay","category":"realtime","also":["video"],"tagline":"Deliver live video with very little delay, using a hosted relay.","description":"MoQ Relay is a hosted service for sending live video and audio to many viewers with very little delay. It uses Media over QUIC (MoQ), a newer internet standard for live media. If you are building an interactive broadcast where viewers react in real time, you can set up your own separate relay and use tokens, which are secure passes, to control who can broadcast and who can watch. It supports MoQ Transport draft-14 and draft-16, with draft-18 in progress. It is in beta and free during the beta.","useCases":["Deliver live video with little delay","Run interactive live broadcasts","Control who can broadcast with tokens"],"pricing":"Free during beta.","status":"beta","url":"https://developers.cloudflare.com/moq/","source":"https://developers.cloudflare.com/moq/"},{"id":"cloudflare-waf","provider":"cloudflare","name":"Cloudflare WAF","category":"waf","tagline":"A shield that blocks common attacks on your website before they land.","description":"Cloudflare WAF is a web application firewall, a filter that checks incoming traffic and blocks attacks before they reach your app. If your site has a login form or a search box, attackers may try tricks like SQL injection, which sneaks database commands into a form, or XSS, which plants harmful scripts. Managed rules catch known attacks for you, and custom rules let you block traffic by country or IP address. Basic managed protection is free, and fuller rule sets come with Pro, Business and Enterprise plans.","useCases":["Block SQL injection and XSS attacks","Block traffic from certain countries","Block known bad IP addresses"],"pricing":"Free managed basics; fuller rulesets on Pro, Business, Enterprise.","status":"ga","url":"https://developers.cloudflare.com/waf/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-rate-limiting","provider":"cloudflare","name":"Rate limiting rules","category":"waf","also":["ddos"],"tagline":"Slow down or block anyone sending too many requests too fast.","description":"Rate limiting rules count how often someone hits your site or API and step in when they go over a limit you choose. If your login page is being hammered by someone guessing passwords, you can block them after a few tries. The same idea works for stopping scrapers or people abusing your API, which is the way other software talks to your app. You can count by IP address or other details of the request, and choose to block, challenge or just log. It is included with your plan, and limits vary by plan.","useCases":["Block password guessing on logins","Limit API abuse from one sender","Slow down scrapers hitting your site"],"pricing":"Included with plan; limits vary by plan.","status":"ga","url":"https://developers.cloudflare.com/waf/rate-limiting-rules/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-api-shield","provider":"cloudflare","name":"API Shield","category":"waf","tagline":"Find, check and lock down the doors other software uses to reach your app.","description":"API Shield protects your APIs, the connections that let apps and other software talk to your service. It finds endpoints, the addresses your API answers at, that you may have forgotten about, and checks that incoming requests match the format you expect. If your mobile app or partners call your API, you can require client certificates (mTLS), a kind of digital ID card, so only trusted callers get in. It also checks login tokens (JWT validation) and spots abuse patterns. It is an Enterprise add-on, with some features on lower plans.","useCases":["Find forgotten API endpoints","Require client certificates from partners","Check API requests match expected format"],"pricing":"Enterprise add-on; some features on lower plans.","status":"ga","url":"https://developers.cloudflare.com/api-shield/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-client-side-security","provider":"cloudflare","name":"Client-side security (Page Shield)","category":"waf","also":["threat-intel"],"tagline":"Watch the scripts and cookies loading in your visitors' browsers.","description":"Client-side security, formerly called Page Shield, watches what code runs in your visitors' browsers. If your checkout page loads scripts from other companies, like analytics or chat widgets, one of them could be changed to steal card details. This tool spots malicious or changed third-party scripts and can alert you or block them using content security policies, which are rules telling browsers what is allowed to load. It also helps with the script lists that payment card (PCI) rules ask for. Basic features are free, and advanced ones need a paid plan.","useCases":["Catch changed third-party scripts on checkout","Stop card-skimming scripts on payment pages","Keep script lists for PCI audits"],"pricing":"Basic free; advanced on paid plans.","status":"ga","url":"https://developers.cloudflare.com/client-side-security/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-bot-management","provider":"cloudflare","name":"Bot Management","category":"bots","tagline":"Spot and stop bad bots, with free basic tools and a deeper paid option.","description":"Bot Management helps you tell real visitors from bots, which are programs that hit your site on their own. If scrapers are copying your content or attackers are testing stolen passwords on your login, it gives each request a bot score and lets you set rules for different pages. Bot Fight Mode is free, and Super Bot Fight Mode comes with Pro and above. The full Enterprise Bot Management, with scores and analytics, is available through Cloudflare sales.","useCases":["Stop scrapers copying your content","Block credential stuffing on logins","Protect checkout from bots"],"pricing":"Bot Fight Mode free; Super Bot Fight on Pro+; Enterprise contact sales.","status":"ga","url":"https://developers.cloudflare.com/bots/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-botbase","provider":"cloudflare","name":"BotBase","category":"bots","also":["threat-intel"],"tagline":"A lookup directory of known bots and AI agents that visit your site.","description":"BotBase is a directory of known bots and agents (automated programs that visit websites), showing how Cloudflare classifies each one. If you spot odd traffic on your site, you can check which bot it is and filter your traffic down to just that one. That makes it easier to write rules aimed at a single bot. If you run a bot yourself, you can submit it to be verified. The full directory view in the dashboard needs Enterprise Bot Management, but a public copy lives on Cloudflare Radar.","useCases":["Identify which bot is visiting","Filter traffic down to one bot","Submit your own bot for verification"],"pricing":"Enterprise Bot Management customers.","status":"ga","url":"https://developers.cloudflare.com/bots/botbase/","source":"https://developers.cloudflare.com/bots/botbase/"},{"id":"cloudflare-turnstile","provider":"cloudflare","name":"Cloudflare Turnstile","category":"bots","tagline":"A free CAPTCHA alternative that skips the picture puzzles.","description":"Turnstile checks that a visitor is a real person without making them solve annoying puzzles. If your app has a sign-up form, a login page or a contact form, you can add it to keep out fake accounts and spam. It runs lightweight challenges that most visitors barely notice. Unlike many Cloudflare tools, it also works on sites that don't use Cloudflare for DNS (your domain's address book) or CDN (fast content delivery). It's free, with Enterprise plans adding extra features.","useCases":["Stop fake sign-ups without puzzles","Protect login forms from bots","Block spam on contact forms"],"pricing":"Free; Enterprise plans with extra features.","status":"ga","url":"https://developers.cloudflare.com/turnstile/","source":"https://developers.cloudflare.com/products/","gem":"A CAPTCHA that never asks anyone to click traffic lights. Free."},{"id":"cloudflare-ddos-protection","provider":"cloudflare","name":"DDoS Protection","category":"ddos","tagline":"Automatic protection against floods of traffic that knock sites offline.","description":"A DDoS attack is when someone floods your site with junk traffic to knock it offline. This service spots and stops those attacks automatically, around the clock, on every plan. If your product suddenly gets attention, good or bad, you don't have to scramble or set anything up. It covers layers 3, 4 and 7, meaning both raw network floods and floods of web requests. You can tune how sensitive it is, and Enterprise adds Advanced options. It's unmetered, so a big attack doesn't change your bill.","useCases":["Stop floods of junk traffic automatically","Keep sites online during sudden attacks","Handle floods of web requests too"],"pricing":"Unmetered and included on all plans.","status":"ga","url":"https://developers.cloudflare.com/ddos-protection/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-account-abuse-protection","provider":"cloudflare","name":"Account Abuse Protection","category":"bots","also":["identity"],"tagline":"Spot fake sign-ups and stolen-password logins, in early access.","description":"Account Abuse Protection helps you find fraud in your login and sign-up traffic, including logins that use leaked credentials (usernames and passwords exposed in data breaches). It builds a running picture of each account's activity from an identifier you configure, uses hashed User IDs (scrambled so the real ID stays private), and includes an investigation dashboard. If fake accounts or account takeovers are a problem, you can use those hashed IDs in WAF rules (website firewall rules). It is in early access, so you'll need to contact sales.","useCases":["Spot fake sign-ups automatically","Catch logins using leaked passwords","Review possible account takeovers in one dashboard"],"pricing":"Early access; contact sales.","status":"beta","url":"https://developers.cloudflare.com/bots/account-abuse-protection/","source":"https://developers.cloudflare.com/bots/account-abuse-protection/"},{"id":"cloudflare-leaked-credentials-detection","provider":"cloudflare","name":"Leaked credentials detection","category":"bots","also":["waf"],"tagline":"Flag logins that use passwords already exposed in known data breaches.","description":"Leaked credentials detection checks whether the login details in a request show up in known data breaches. It compares hashed versions (scrambled copies), so the real passwords stay private. If your app has a login page, it helps you catch people reusing a password that's already out in the wild, or attackers trying stolen lists. You can block those requests or treat them as higher risk. It's free on all plans, so it's an easy first layer for login safety.","useCases":["Block logins with passwords exposed in breaches","Flag reused passwords already found online","Score login risk for suspicious requests"],"pricing":"Free on all plans.","status":"ga","url":"https://developers.cloudflare.com/waf/detections/leaked-credentials/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-fraud-detection","provider":"cloudflare","name":"Fraud Detection (sequence rules)","category":"bots","also":["waf"],"tagline":"Catch abuse by watching the order and timing of a visitor's requests.","description":"Sequence rules use cookies (small notes stored in the visitor's browser) to track the order and timing of requests during a session. If you run a checkout, a real shopper moves through steps in a sensible order, while an attacker might skip a step or rush through. Sequence rules can flag that pattern. This is different from rules that judge each request alone, so it catches abuse that looks normal one request at a time. It's an Enterprise API Shield and Bot Management feature.","useCases":["Flag checkouts where steps are skipped","Spot shoppers moving through steps too fast","Catch abuse that looks normal per request"],"pricing":"Enterprise API Shield / Bot Management feature.","status":"ga","url":"https://developers.cloudflare.com/waf/detections/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-challenges","provider":"cloudflare","name":"Challenges","category":"bots","also":["waf"],"tagline":"Quick checks that tell real visitors from bots.","description":"Challenges are short checks your site can show suspicious visitors to confirm they're not bots. Cloudflare offers managed, JS and interactive challenges, and your WAF (website firewall) and Bots rules can trigger them. If a login page is getting strange traffic, you can challenge those visitors instead of blocking everyone. Real people usually pass quickly, and you rely less on traditional CAPTCHA puzzles. Challenges come included with your plan, so there's nothing extra to buy.","useCases":["Show quick checks to suspicious visitors","Challenge bots on a busy login page","Use fewer traditional CAPTCHA puzzles"],"pricing":"Included with plan.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-challenges/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-privacy-pass","provider":"cloudflare","name":"Privacy Pass","category":"bots","also":["identity"],"tagline":"Let real visitors skip repeat bot checks without being tracked.","description":"Privacy Pass lets a visitor prove something about themselves, such as passing a bot check, using a token (a small digital proof) that doesn't reveal who they are. If your site challenges visitors, this means people don't have to solve a CAPTCHA over and over. Cloudflare implements the Privacy Pass protocol, so users get fewer interruptions without being tracked across sites. Pick it when you care about both blocking bots and respecting visitor privacy. The protocol itself is free.","useCases":["Let real visitors skip repeated bot checks","Cut down on repeat CAPTCHA puzzles","Prove a visitor passed a check privately"],"pricing":"Free protocol.","status":"ga","url":"https://developers.cloudflare.com/privacy-pass/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-access","provider":"cloudflare","name":"Cloudflare Access","category":"zero-trust","also":["identity"],"tagline":"Put a login in front of private apps, no VPN needed.","description":"Cloudflare Access checks who someone is, using your identity provider (the service that handles your team's logins), before letting them into an app. It also logs every request. If you have an admin panel, an internal dashboard or contractors who need temporary access, you can protect them without a VPN (a private network tunnel). It works for self-hosted apps, SaaS tools, SSH and other infrastructure. It's free for up to 50 users, with per-user pricing above that.","useCases":["Put a login in front of admin panels","Give contractors temporary access without a VPN","Replace VPNs for internal tools"],"pricing":"Free for up to 50 users; per-user pricing above.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/access-controls/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-gateway","provider":"cloudflare","name":"Cloudflare Gateway","category":"zero-trust","also":["dns"],"tagline":"Filter and inspect your team's internet traffic to block threats.","description":"Cloudflare Gateway is a checkpoint for the internet traffic leaving your team's devices. It applies policies to DNS lookups, network traffic, web requests and outbound connections. If you want to block known malware websites or inspect HTTPS traffic (the encrypted kind) for threats, you set the rules once and they apply everywhere. It can also give you dedicated egress IPs, fixed addresses your traffic comes from, which other services can then allow. It's free for up to 50 users, with per-user pricing above.","useCases":["Block known malware websites for the team","Inspect encrypted web traffic for threats","Give traffic fixed outgoing IP addresses"],"pricing":"Free for up to 50 users; per-user above.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/traffic-policies/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloudflare-one-client","provider":"cloudflare","name":"Cloudflare One Client (WARP)","category":"zero-trust","also":["networking"],"tagline":"An app for your team's devices that connects them safely to Cloudflare.","description":"The Cloudflare One Client is a small app you install on laptops and phones. It sends their traffic to Cloudflare and reports device health, such as whether a device meets your security rules. If your team works from cafes or home, it keeps Gateway protection on wherever they are and gives them access to private networks. It was formerly called the WARP client, and the separate consumer WARP app is a different product. It's part of Cloudflare One per-user pricing.","useCases":["Keep team laptops protected on any network","Connect remote devices to private networks","Check whether devices meet security rules"],"pricing":"Part of Cloudflare One per-user pricing.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/team-and-resources/devices/cloudflare-one-client/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-dlp","provider":"cloudflare","name":"Data Loss Prevention","category":"zero-trust","tagline":"Spot and block sensitive data before it leaves through web or SaaS apps.","description":"Data Loss Prevention scans web traffic and SaaS apps (online tools your team uses) for sensitive information, then detects and blocks it from moving where it shouldn't. It uses profiles, validators and classification to recognize what counts as sensitive. If someone might paste customer details or source code into the wrong place, it can catch PII (personal information) and code leaks. It's a Cloudflare One add-on, so you'll need to contact sales.","useCases":["Detect personal information leaving through apps","Block source code pasted into wrong places","Catch customer details sent to outside tools"],"pricing":"Cloudflare One add-on; contact sales.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/data-loss-prevention/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-casb","provider":"cloudflare","name":"CASB","category":"zero-trust","tagline":"Scan your SaaS and cloud accounts for risky settings and exposed files.","description":"CASB (cloud access security broker) checks the online tools your team uses, like file storage and business apps, for risky settings and sensitive content. If your company keeps customer documents in a shared drive, it can flag files that are open to anyone with a link. It connects to your accounts through their APIs (the built-in way apps talk to each other), so you don't have to install anything on laptops. Pick it when you want to find mistakes in the tools you already use. It is an add-on to Cloudflare One, and you contact sales for pricing.","useCases":["Find risky settings in SaaS accounts","Flag files open to anyone with a link","Check business apps without installing software"],"pricing":"Cloudflare One add-on; contact sales.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/integrations/cloud-and-saas/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-browser-isolation","provider":"cloudflare","name":"Browser Isolation","category":"zero-trust","tagline":"Open risky websites in a safe remote browser, away from your device.","description":"Browser Isolation opens web pages in a sealed-off browser running somewhere else instead of on your own computer, so harmful code never touches the device. If you work with contractors on their own laptops, or your team has to visit unfamiliar sites, the page runs remotely and the person only sees the result. That cuts the risk from malware (harmful software) and phishing (fake pages that steal logins). Pick it over a normal browser when you cannot fully trust the device or the site. It is an add-on to Cloudflare One.","useCases":["Open risky sites in a remote browser","Keep malware away from contractor devices","Block phishing pages from stealing logins"],"pricing":"Cloudflare One add-on.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/remote-browser-isolation/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-email-security","provider":"cloudflare","name":"Email Security","category":"email-security","also":["zero-trust"],"tagline":"Catch phishing, malware and fake-sender emails before they reach your team.","description":"Email Security scans your email for phishing (messages that trick people into giving up logins), malware, and impersonation, where someone pretends to be a person or brand you trust. If a fake message that looks like it comes from your CEO asks your finance person to send a payment, this is the kind of attack it is built to catch. It covers both incoming and outgoing mail, with PhishGuard, its phishing protection. It was formerly called Area 1. Pricing is through sales.","useCases":["Catch phishing emails before the team sees them","Block fake emails impersonating your executives","Scan incoming and outgoing email for malware"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/email-security/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-dmarc-management","provider":"cloudflare","name":"DMARC Management","category":"email-security","also":["email"],"tagline":"See who is sending email as your domain and shut down spoofers.","description":"DMARC Management gathers the reports that email providers send about messages using your domain name and turns them into easy charts. DMARC is a setting that tells inbox providers what to do with fake emails pretending to be you. If you are launching a product and want customers to trust your emails, it helps you spot spoofing and move step by step to strict enforcement, where fakes get blocked. It is free, which makes it an easy place to start.","useCases":["See who sends email as your domain","Roll out strict email rules step by step","Spot spoofed emails pretending to be you"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/dmarc-management/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloud-workload-security","provider":"cloudflare","name":"Cloud Workload Security","category":"zero-trust","also":["management"],"tagline":"Set rules that check and block risky requests to your cloud accounts.","description":"Cloud Workload Security sits in the path of the requests your apps or AI agents send to cloud services and checks each one against rules you write. If an agent tries to make a storage bucket (a cloud folder for files) public, a rule can stop that change before it happens. Rules are written in Rego, a policy language, and you set the scope with workloads, policy sets, deployments and domain access lists. Pick it when you want guardrails on what automated tools can do. It is in beta, and pricing is not stated.","useCases":["Block AI agents from making storage public","Write rules for automated cloud requests","Check requests sent to cloud services"],"pricing":"Not stated in docs.","status":"beta","url":"https://developers.cloudflare.com/cloudflare-one/cloud-workload-security/","source":"https://developers.cloudflare.com/cloudflare-one/cloud-workload-security/"},{"id":"cloudflare-dex","provider":"cloudflare","name":"Digital Experience Monitoring","category":"observability","also":["zero-trust"],"tagline":"Spot why a laptop, network or app is slow or failing for your team.","description":"Digital Experience Monitoring runs tests on your team's enrolled devices and collects numbers on how the device, the network and the apps are doing. If an employee says the internet is broken, you can see whether the problem is their laptop, their Wi-Fi, or their internet provider. It is a good pick when you have a remote team and need real answers faster than guessing. It is a Cloudflare One feature, and what you get depends on your plan.","useCases":["Find whether a laptop or Wi-Fi is slow","Tell if your internet provider is at fault","Diagnose team device problems faster than guessing"],"pricing":"Cloudflare One feature; plan dependent.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/insights/dex/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-tunnel","provider":"cloudflare","name":"Cloudflare Tunnel","category":"networking","also":["zero-trust"],"tagline":"Put a local app or server on the internet safely, with no open ports.","description":"Cloudflare Tunnel connects your server to Cloudflare by making an outgoing connection from your side, so you never have to open the door to incoming traffic. If you built a prototype on your laptop and want to show a client, you can share it at a web address without exposing your machine directly. It runs through cloudflared, a small program you install on the server, and works for both public sites and private tools. Pick it over opening ports yourself when you want fewer things to lock down. It is free.","useCases":["Share a local prototype at a web address","Show a client a laptop prototype","Expose a server without opening ports"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-one/networks/connectors/cloudflare-tunnel/","source":"https://developers.cloudflare.com/products/","gem":"Put a server online with no open ports and no public IP."},{"id":"cloudflare-mesh","provider":"cloudflare","name":"Cloudflare Mesh","category":"networking","also":["zero-trust"],"tagline":"A private, encrypted network linking your servers, devices and clouds.","description":"Cloudflare Mesh creates a private network where each enrolled machine gets its own private IP address (its internal network number), so your servers, clouds and sites can reach each other as if they were in one place. If your app runs on two different cloud providers and a laptop needs to reach both, Mesh links them. Its encryption is post-quantum, meaning it is built to hold up against future quantum computers. It replaces WARP Connector and peer-to-peer connectivity. Release stage and pricing are not stated.","useCases":["Link servers, devices and clouds privately","Let a laptop reach two clouds","Keep links encrypted against future quantum attacks"],"pricing":"Not stated in docs.","status":"ga","url":"https://developers.cloudflare.com/mesh/","source":"https://developers.cloudflare.com/mesh/"},{"id":"cloudflare-cloudflare-wan","provider":"cloudflare","name":"Cloudflare WAN (Magic WAN)","category":"networking","tagline":"Connect your offices and data centers through Cloudflare's network.","description":"Cloudflare WAN, also known as Magic WAN, joins your offices and data centers into one private network that runs over Cloudflare's network. A WAN (wide area network) is simply the link between separate sites. If your company has several branches that need to share internal tools, you can connect them using GRE or IPsec tunnels (two standard ways to link networks) or a direct interconnect, with firewall and routing included. Pick it when you want to replace older WAN or SD-WAN setups. Pricing is through sales.","useCases":["Connect branch offices into one network","Link data centers through Cloudflare's network","Replace older office network setups"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-wan/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloudflare-one-appliance","provider":"cloudflare","name":"Cloudflare One Appliance","category":"networking","also":["zero-trust"],"tagline":"A managed box, physical or virtual, that connects a branch to Cloudflare.","description":"Cloudflare One Appliance is a managed device, either physical hardware or a virtual version (software running on a server), that you place at a branch site to connect it to Cloudflare One. If you open a new office and want it on your company network without building the setup by hand, this is the on-ramp. It is a good choice when you would rather have a managed device than assemble your own. Pricing is through sales.","useCases":["Connect a new branch office quickly","Put a managed device at each branch","Get a managed device instead of DIY setup"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-wan/configuration/appliance/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-network-firewall","provider":"cloudflare","name":"Cloudflare Network Firewall","category":"networking","also":["ddos"],"tagline":"Firewall rules for your company network, run as a service by Cloudflare.","description":"Cloudflare Network Firewall lets you set rules about which traffic can enter your company network, without buying and running a firewall box yourself. A firewall is a gatekeeper that blocks unwanted connections. If you want only certain IP addresses (the numeric addresses of computers) to reach an internal system, you write a rule here. It applies to traffic entering through Magic Transit and Cloudflare WAN, so it fits best if you already use one of them. Pricing is through sales.","useCases":["Allow only certain addresses into internal systems","Write firewall rules without running hardware","Filter traffic entering your company network"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/magic-firewall/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-magic-transit","provider":"cloudflare","name":"Magic Transit","category":"networking","also":["ddos"],"tagline":"DDoS protection and faster traffic for your whole block of IP addresses.","description":"Magic Transit protects entire networks, not just a website, from DDoS attacks, which flood you with fake traffic to knock you offline. Cloudflare announces your IP prefixes (your block of internet addresses) through its network, so traffic arrives there first and gets cleaned before it reaches you. If you run your own data center, or a mix of cloud and on-site servers, it shields them all. Pick it over website-only protection when the thing you must defend is the network itself. Pricing is through sales.","useCases":["Shield whole data centers from floods","Protect your full block of IP addresses","Clean traffic before it reaches your servers"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/magic-transit/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-network-flow","provider":"cloudflare","name":"Network Flow","category":"observability","also":["networking"],"tagline":"See what traffic crosses your network and spot attacks early.","description":"Network Flow, formerly called Magic Network Monitoring, shows you what traffic is moving across your network and your cloud. It reads flow data, which is a summary of who is talking to whom and how much, and looks for oddities and signs of a DDoS attack (a flood of junk traffic meant to knock you offline). If you run your own servers or a company network and need to know why traffic suddenly spiked, this is where you would look. Pick it when you want clear visibility and early warning. It is sold through Cloudflare's sales team, and free for some accounts.","useCases":["See what traffic crosses your network","Spot DDoS attacks early in traffic","Explain sudden traffic spikes quickly"],"pricing":"Contact sales; free for some accounts.","status":"ga","url":"https://developers.cloudflare.com/network-flow/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-network-interconnect","provider":"cloudflare","name":"Network Interconnect","category":"networking","tagline":"A private link from your network straight to Cloudflare.","description":"Network Interconnect gives you a direct physical or virtual link between your own network and Cloudflare, instead of sending traffic over the public internet. If you run your own data center and push a lot of traffic to Cloudflare, a private link gives that traffic its own road. These links are faster and more secure than ordinary internet paths. Pick it when you have serious network needs and want a dedicated on-ramp rather than the shared internet. Pricing is handled through Cloudflare's sales team.","useCases":["Link your network directly to Cloudflare","Give heavy traffic a private route","Move data center traffic off public internet"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/network-interconnect/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-byoip","provider":"cloudflare","name":"BYOIP and Address maps","category":"networking","tagline":"Use your own IP addresses with Cloudflare, and pin hostnames to them.","description":"BYOIP, short for bring your own IP, lets you run Cloudflare services on IP address ranges your company already owns. Address maps let you tie specific hostnames (like app.yourcompany.com) to specific IP addresses. If a partner or bank will only accept traffic from a fixed list of IP addresses, or you want your brand's own ranges to show up, this is how you keep them. Pick it over Cloudflare's default addresses when the exact IPs matter to someone else. It is an Enterprise product, so you talk to sales.","useCases":["Run Cloudflare on your own IP ranges","Give partners a fixed IP allowlist","Show your brand's own IP ranges"],"pricing":"Enterprise; contact sales.","status":"ga","url":"https://developers.cloudflare.com/byoip/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-spectrum","provider":"cloudflare","name":"Spectrum","category":"ddos","also":["networking"],"tagline":"DDoS protection for game servers, SSH, mail and other non-web traffic.","description":"Spectrum puts Cloudflare's protection and speed in front of apps that are not ordinary websites. It works with TCP and UDP, the two basic ways computers send data, so it covers things like SSH (remote server logins), game servers and mail. If you host a multiplayer game and worry about attackers flooding it with junk traffic, Spectrum sits in front and absorbs the attack. Pick it when your product speaks something other than web pages. It is an Enterprise product, though some plans include limited access.","useCases":["Protect a multiplayer game server","Shield remote SSH logins from attacks","Guard mail and other non-web traffic"],"pricing":"Enterprise; some plans offer limited access.","status":"ga","url":"https://developers.cloudflare.com/spectrum/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-load-balancing","provider":"cloudflare","name":"Load Balancing","category":"load-balancing","tagline":"Spread traffic across servers and regions, with automatic failover.","description":"Load Balancing splits incoming traffic across several servers or regions and checks that each one is healthy. If your app has servers in two regions and one goes down, traffic moves to the other one automatically. You can also send visitors to the closest region, which is called geo steering. Pick it when one server is not enough or you cannot afford downtime. It is an add-on, priced by the number of origins (your servers) and the health checks you run.","useCases":["Switch to another region when one fails","Send visitors to the nearest region","Share traffic across several servers"],"pricing":"Add-on; priced per origin and checks.","status":"ga","url":"https://developers.cloudflare.com/load-balancing/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-health-checks","provider":"cloudflare","name":"Health Checks","category":"load-balancing","also":["observability"],"tagline":"Get alerted when your server or website stops responding.","description":"Health Checks keep an eye on your server's IP address or hostname and notify you when something breaks. They test from several locations around the world, so you can tell a real outage from a local hiccup. If your online store goes down at 2 a.m., you hear about it before your customers do. Pick it when you want simple uptime alerts without setting up a separate monitoring tool. It is available on the Pro plan or higher.","useCases":["Get alerts when your site stops responding","Confirm a real outage, not a local hiccup","Monitor uptime without a separate tool"],"pricing":"Pro plan or higher.","status":"ga","url":"https://developers.cloudflare.com/health-checks/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-waiting-room","provider":"cloudflare","name":"Waiting Room","category":"load-balancing","tagline":"A virtual queue that keeps your site standing when a crowd shows up.","description":"Waiting Room puts visitors into an orderly online line when more people arrive than your site can handle. Instead of crashing, your site lets people in fairly, a bit at a time, while everyone else waits. If you are selling concert tickets or running a limited product drop, this protects both your server and your launch day. Pick it when you expect sudden spikes and would rather queue people than lose them to errors. It is an add-on for Business and Enterprise plans.","useCases":["Queue fans for concert ticket sales","Manage crowds during limited product drops","Keep your server up during traffic spikes"],"pricing":"Business and Enterprise add-on.","status":"ga","url":"https://developers.cloudflare.com/waiting-room/","source":"https://developers.cloudflare.com/products/","gem":"A polite virtual line when a launch sends more traffic than you can take."},{"id":"cloudflare-argo-smart-routing","provider":"cloudflare","name":"Argo Smart Routing","category":"cdn","tagline":"Send traffic along the fastest, least crowded network routes.","description":"Argo Smart Routing picks the quickest path for traffic traveling between Cloudflare and your server. It uses live network measurements to steer around congested routes, much like a navigation app avoiding a traffic jam. If your visitors are far from your server and pages feel slow or sometimes fail, it can lower delays and cut errors. Pick it when speed and reliability matter more than saving a few dollars. It is an add-on billed by usage, per GB of data.","useCases":["Speed up pages for faraway visitors","Route around congested network paths","Reduce failed requests and timeouts"],"pricing":"Per-GB usage add-on.","status":"ga","url":"https://developers.cloudflare.com/argo-smart-routing/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-smart-shield","provider":"cloudflare","name":"Smart Shield","category":"cdn","also":["ddos"],"tagline":"Protect your server from too many requests with caching and shared connections.","description":"Smart Shield is a bundle that reduces how often visitors' requests actually reach your own server (your origin). It does this with Smart Tiered Cache and connection reuse, which means fewer repeat trips and fewer new connections. If your app gets a sudden rush of visitors, your server is not left to cope alone. Higher tiers add extras such as Argo, Regional Tiered Cache, Cache Reserve and Dedicated CDN Egress IPs. Pick it when you want one place to manage origin protection. Pricing depends on the tier.","useCases":["Shield your server from request rushes","Cut requests reaching your own server","Manage server protection in one place"],"pricing":"Depends on package tier.","status":"ga","url":"https://developers.cloudflare.com/smart-shield/","source":"https://developers.cloudflare.com/smart-shield/"},{"id":"cloudflare-tiered-cache","provider":"cloudflare","name":"Tiered Cache","category":"cdn","tagline":"Cut repeat requests to your server by having fewer Cloudflare locations ask for files.","description":"Tiered Cache means only a few upper-tier Cloudflare data centers contact your server for files they do not already have. Other locations ask those data centers first. If your site serves the same images or pages to visitors worldwide, your server answers far fewer requests. It is now set up through Smart Shield. Pick it when your server feels the load of lots of cache misses. Smart Tiered Cache is free, while Regional Tiered Cache is on Enterprise.","useCases":["Reduce repeat file requests to your server","Lighten load from lots of cache misses","Serve worldwide images with fewer origin hits"],"pricing":"Free for Smart Tiered Cache; Regional on Enterprise.","status":"ga","url":"https://developers.cloudflare.com/cache/how-to/tiered-cache/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cache-reserve","provider":"cloudflare","name":"Cache Reserve","category":"cdn","also":["object-storage"],"tagline":"Keep rarely requested files cached longer so your server is asked less.","description":"Cache Reserve stores your cacheable files in R2-backed storage, which is Cloudflare's own file storage, so they stay cached much longer. Normally rarely requested files get dropped from the cache and have to be fetched from your server again. If you serve a large library of images or downloads that people open only now and then, Cache Reserve keeps them close. Pick it when you want to save on data your server sends out. You pay per GB stored and for operations.","useCases":["Keep rarely downloaded files cached longer","Cut data your server sends out","Store large image or download libraries"],"pricing":"Per GB stored and operations.","status":"ga","url":"https://developers.cloudflare.com/cache/advanced-configuration/cache-reserve/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cdn-cache","provider":"cloudflare","name":"CDN and Cache Rules","category":"cdn","tagline":"Serve your site's files from locations around the world, with control over caching.","description":"The CDN (content delivery network) keeps copies of your site's files in Cloudflare data centers around the world, so visitors load them from somewhere nearby. Cache Rules let you decide what gets saved and for how long, and purge clears out old copies. Crawler Hints and Always Online are also included. If your app serves images, scripts and styles to a global audience, this is the easy first step to faster pages. It is free on all plans.","useCases":["Serve images and scripts from nearby locations","Set how long files stay cached","Clear out old cached copies"],"pricing":"Free on all plans.","status":"ga","url":"https://developers.cloudflare.com/cache/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-china-network","provider":"cloudflare","name":"China Network","category":"cdn","tagline":"Serve visitors inside and outside China with one network.","description":"China Network is a Cloudflare option that helps your website or app load quickly for visitors in Mainland China, using a partnership with JD Cloud, a large Chinese cloud company. If you sell to customers both in China and elsewhere, you can serve them all through one setup instead of running a separate system for each region. It is worth a look when your site feels slow for visitors in China and you want to keep the rest of your setup with Cloudflare. It is priced as an Enterprise plan.","useCases":["Load your site faster for China visitors","Serve China and global customers together","Keep one setup for all regions"],"pricing":"Enterprise.","status":"ga","url":"https://developers.cloudflare.com/china-network/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-apo","provider":"cloudflare","name":"Automatic Platform Optimization","category":"cdn","also":["hosting"],"tagline":"Make WordPress sites load faster by serving pages from Cloudflare.","description":"Automatic Platform Optimization, or APO, stores full copies of your WordPress pages on Cloudflare's servers around the world, so visitors get them from somewhere close by. If you run a blog or online shop on WordPress and pages feel slow, this can speed things up without rebuilding anything. Ordinary caching usually covers images and style files, while APO saves the whole page. It comes as a paid add-on or with the Pro plan and above.","useCases":["Speed up slow WordPress blog pages","Cache full WordPress pages worldwide","Speed up WordPress shops without rebuilding"],"pricing":"Paid add-on or Pro+.","status":"ga","url":"https://developers.cloudflare.com/automatic-platform-optimization/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-snippets","provider":"cloudflare","name":"Snippets","category":"edge-compute","tagline":"Small bits of code that tweak requests and responses as they pass.","description":"Snippets are small pieces of JavaScript, a common web programming language, that run on Cloudflare's network as visitors reach your site. If you need to change a header, a small piece of information sent along with each page, or send certain visitors somewhere else, a snippet can do it in a few lines. Pick Snippets over a full Worker, Cloudflare's tool for building whole apps, when the job is a quick tweak. They are included on paid plans.","useCases":["Change a header on each request","Redirect certain visitors with a few lines","Tweak responses without a full app"],"pricing":"Included on paid plans.","status":"ga","url":"https://developers.cloudflare.com/rules/snippets/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloud-connector","provider":"cloudflare","name":"Cloud Connector","category":"cdn","also":["object-storage"],"tagline":"Point your domain at a storage bucket to host a static site.","description":"Cloud Connector is a rule that sends matching visitor requests straight to a storage bucket, which is an online folder for files, such as Amazon S3, Google Cloud Storage or Cloudflare R2. If you built a static site, one made of plain files with no server behind it, you can keep it in a bucket and let your domain serve it. It saves you from setting up a separate server just to hand out files. It is included at no extra charge.","useCases":["Host a static site from a storage bucket","Point your domain at stored files","Skip a separate server for plain files"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/rules/cloud-connector/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-rules","provider":"cloudflare","name":"Rules (Redirect, Transform, Origin)","category":"cdn","tagline":"Redirect visitors and rewrite requests with simple rules.","description":"Rules is Cloudflare's set of tools for changing how requests to your site are handled. It includes Redirect Rules, Transform Rules, which rewrite parts of a request or response such as headers, Managed Transforms, and a simulator to test rules before they go live. If you want old links to land on new pages or need to add a header, you can do it here without touching your app's code. Choose Rules over the older Page Rules for new setups. It is included with your plan.","useCases":["Redirect old links to new pages","Add or rewrite headers without code changes","Test rules with a simulator first"],"pricing":"Included with plan.","status":"ga","url":"https://developers.cloudflare.com/rules/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-bulk-redirects","provider":"cloudflare","name":"Bulk Redirects","category":"cdn","tagline":"Redirect thousands of old links to new ones using one uploaded list.","description":"Bulk Redirects lets you upload a long list of old web addresses and where each one should go, then switch them all on with a single rule. If you are moving to a new site design or domain and hundreds of old links need to keep working, this saves you from creating each redirect by hand. Choose it over one-off redirect rules when your list is big. It is included with your plan, up to the plan's limits.","useCases":["Move hundreds of old links at once","Upload one list of redirects","Keep old links working after a migration"],"pricing":"Included with plan limits.","status":"ga","url":"https://developers.cloudflare.com/rules/url-forwarding/bulk-redirects/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-page-rules","provider":"cloudflare","name":"Page Rules","category":"cdn","tagline":"Older per-page settings, now replaced by Rules.","description":"Page Rules lets you set how Cloudflare treats certain web addresses on your site, such as redirecting one page or changing how long it is saved for faster loading. You will mostly meet it in older setups, since it is still in the dashboard so existing configurations keep working. If you are starting fresh, use Rules instead, which is the newer replacement. Page Rules is deprecated, which means Cloudflare has replaced it and you should not build anything new on it.","useCases":["Keep legacy redirects in older setups","Maintain existing per-page settings","Review old settings before moving to Rules"],"pricing":"Included on plans.","status":"deprecated","url":"https://developers.cloudflare.com/rules/page-rules/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-custom-error-pages","provider":"cloudflare","name":"Custom Error Pages","category":"cdn","tagline":"Show your own branded page when something goes wrong.","description":"Custom Error Pages let you replace the plain default messages visitors see when your site hits a problem with pages that look like your brand. If your server goes down or Cloudflare can't reach it, visitors can see a friendly page with your logo and a way to contact you instead of a cold technical message. It covers errors from your own server and from Cloudflare itself. It is included with your plan.","useCases":["Show branded pages when your server fails","Offer a friendly contact link during outages","Replace plain error messages with your logo"],"pricing":"Included with plan.","status":"ga","url":"https://developers.cloudflare.com/rules/custom-error-responses/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-network-settings","provider":"cloudflare","name":"Network settings","category":"networking","tagline":"Switch on IPv6, WebSockets, gRPC and other connection options.","description":"Network settings is a page of on and off switches for how visitors connect to your site through Cloudflare. It covers IPv6, the newer style of internet address, WebSockets, which keep a live two-way connection open for things like chat, and gRPC, a way for apps to talk to each other. It also includes True-Client-IP, a label that passes along the visitor's real address. If you are building a live-update feature, this is where you make sure WebSockets are turned on. It is included.","useCases":["Turn on WebSockets for live chat","Enable IPv6 for newer internet addresses","Allow gRPC connections between apps"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/network/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-onion-routing","provider":"cloudflare","name":"Onion Routing","category":"networking","also":["waf"],"tagline":"Let Tor browser users reach your site through a private .onion address.","description":"Onion Routing gives your site a Cloudflare-managed .onion address, a special web address that people using the Tor browser, a privacy-focused way to browse, can visit. If some of your visitors protect their privacy with Tor, they can run into extra security challenges on regular sites. This feature reduces those challenges. Turn it on if you want your site to be friendlier to privacy-minded visitors. It is free.","useCases":["Reach Tor browser visitors through a .onion address","Reduce security challenges for privacy-minded visitors","Make your site friendlier to Tor users"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/network/onion-routing/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-network-error-logging","provider":"cloudflare","name":"Network Error Logging","category":"observability","also":["networking"],"tagline":"See when visitors can't connect to your site, and where.","description":"Network Error Logging collects reports from your visitors' browsers when they fail to connect to your site. It shows when and where the problems happen and how much they affect people. If customers say your site is down but everything looks fine on your end, this can show whether the trouble sits with a particular internet provider. It is included, so you don't pay extra to turn it on.","useCases":["Find where visitors fail to connect","Spot problems tied to specific internet providers","Check reports when customers say it's down"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/network-error-logging/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-zaraz","provider":"cloudflare","name":"Zaraz","category":"analytics","also":["edge-compute"],"tagline":"Run your analytics and ad tags without slowing your pages.","description":"Zaraz is a tool for managing third-party tags, the little scripts that add things like analytics and ad tracking to your site. Instead of loading them in your visitors' browsers, it runs them on Cloudflare's network, which helps pages load faster and keeps more visitor data private. If your marketing team keeps adding tracking tools and your site is getting heavier, Zaraz gives you one place to manage them, including consent controls. It has a free tier, then pricing is based on usage.","useCases":["Manage analytics and ad tags in one place","Run tracking scripts without slowing pages","Control consent for visitor tracking"],"pricing":"Free tier, then usage-based.","status":"ga","url":"https://developers.cloudflare.com/zaraz/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-google-tag-gateway","provider":"cloudflare","name":"Google tag gateway for advertisers","category":"analytics","tagline":"Load your Google ad and analytics tags from your own domain.","description":"Google tag gateway makes the Google tracking code on your site, delivered through Google Tag Manager, load from your own domain instead of a Google one. If you run Google ads and want to measure sign-ups or purchases, this keeps the tracking looking like part of your own site, with extra privacy controls added. Pick it when you already use Google Tag Manager and want it served first-party, meaning from your own web address, without rebuilding your setup. It is included with Cloudflare at no extra charge.","useCases":["Load Google tags from your own domain","Keep Google ad tracking under your domain","Measure sign-ups and purchases from Google ads"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/google-tag-gateway/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-web-analytics","provider":"cloudflare","name":"Cloudflare Web Analytics","category":"analytics","tagline":"See visitors and page speed without cookies or tracking your users.","description":"Cloudflare Web Analytics shows how many people visit your site, which pages they view, and how fast those pages load, all without cookies (small files that follow visitors around). If you just launched a landing page and want to know whether your announcement sent traffic, this gives you the numbers. It collects data either with a small snippet of code you add, called a beacon, or straight from Cloudflare's network if your site already runs through it. It is free, and a good pick over heavier tools when you want simple, privacy-friendly stats.","useCases":["Count visitors and top pages privately","Check whether a launch page drew traffic","Track page speed without cookies"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/web-analytics/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-rum-observatory","provider":"cloudflare","name":"Observatory and RUM","category":"observability","also":["analytics"],"tagline":"Test your site's speed and see how it feels for real visitors.","description":"Observatory and RUM live in the Speed tab of your Cloudflare dashboard. Observatory runs speed tests on your site, while RUM (real user monitoring) measures how fast pages actually load for your visitors. Origin Analytics is also there, and it looks at how your own server is doing. If you redesigned your homepage and worry it got slower, you can catch the slowdown and track Core Web Vitals, Google's page speed scores, over time. It is included with your Cloudflare plan.","useCases":["Run speed tests on your site","Measure how fast pages load for visitors","Track Core Web Vitals over time"],"pricing":"Included with plan.","status":"ga","url":"https://developers.cloudflare.com/speed/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-synthetic-monitoring","provider":"cloudflare","name":"Synthetic monitoring","category":"observability","tagline":"Scheduled test visits that check your site is up and fast.","description":"Synthetic monitoring sends automatic test visits to your pages and endpoints (the web addresses your app's backend answers on) on a schedule, from Cloudflare locations around the world. If your sign-up page matters to your business, you can learn it broke before a customer tells you. It also helps you record a baseline, a normal speed to compare against later. Pick it over waiting for real visitors to hit problems, since the tests run even when nobody is on your site. Pricing is not stated, so check the docs.","useCases":["Test sign-up page on a schedule","Catch outages before customers report them","Record a speed baseline for comparison"],"pricing":"Not stated; see docs.","status":"ga","url":"https://developers.cloudflare.com/speed/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-logpush","provider":"cloudflare","name":"Logpush","category":"observability","tagline":"Send your Cloudflare activity logs to storage or security tools.","description":"Logpush automatically sends logs, the detailed records of requests and events on your Cloudflare setup, to places you choose, such as R2 or S3 (file storage), or Datadog and Splunk (monitoring tools). If your company needs a long-term archive for audits, or feeds a SIEM (a security tool that watches for threats), this is how the data gets there. Choose it when you want logs in your own tools rather than only viewing them inside Cloudflare. It is an Enterprise feature, though other plans can send logs to R2.","useCases":["Send logs to security tools","Archive logs for long-term audits","Feed logs into Datadog or Splunk"],"pricing":"Enterprise; R2 destination on others.","status":"ga","url":"https://developers.cloudflare.com/logs/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-log-explorer","provider":"cloudflare","name":"Log Explorer","category":"observability","tagline":"Keep your Cloudflare logs in the dashboard and search them with SQL.","description":"Log Explorer stores your Cloudflare logs inside Cloudflare and lets you search them with SQL, a common language for asking questions of data. If your site saw a strange spike in traffic last night, you can dig through the records to see what happened without setting up another tool. Choose it over Logpush when you want quick answers in one place instead of shipping logs elsewhere first. It is a paid add-on.","useCases":["Search site logs with SQL","Investigate a sudden traffic spike","Get quick answers without extra tools"],"pricing":"Paid add-on.","status":"ga","url":"https://developers.cloudflare.com/log-explorer/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-workers-observability","provider":"cloudflare","name":"Workers Observability","category":"observability","also":["serverless"],"tagline":"See logs, traces and errors for your Cloudflare Workers apps.","description":"Workers Observability shows you what your Workers, small programs that run on Cloudflare's network, are doing. You get logs (notes your code writes), traces (the path one request took) and a query builder to search them, plus a list of issues. If a page in your app feels slow or errors out, you can find the exact request behind it. It can also send data to other tools through OpenTelemetry, a common format for monitoring data, and it is included with usage limits.","useCases":["Find logs behind slow pages","Trace the path of one request","Search app notes and errors together"],"pricing":"Included with usage limits.","status":"ga","url":"https://developers.cloudflare.com/workers/observability/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-opentelemetry-export","provider":"cloudflare","name":"OpenTelemetry export","category":"observability","tagline":"Send your Workers' monitoring data to tools like Honeycomb or Grafana.","description":"OpenTelemetry export sends traces and logs from your Cloudflare Workers (small programs running on Cloudflare's network) to other monitoring tools that accept OTLP, a common format for this kind of data. If your team already watches everything in Honeycomb or Grafana, you can see your Workers there alongside the rest of your app. Pick it when you have a monitoring tool you like and do not want to check a second dashboard. It is included at no extra cost.","useCases":["Send Workers data to Honeycomb or Grafana","Watch Workers beside your other apps","Use your existing monitoring dashboard"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/workers/observability/exporting-opentelemetry-data/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-notifications","provider":"cloudflare","name":"Notifications and Alerts","category":"observability","also":["management"],"tagline":"Get alerted by email or webhook when something happens on your site.","description":"Notifications and Alerts tells you when something important happens on your Cloudflare account. You pick the alerts you want, such as an outage or a DDoS attack (a flood of fake traffic meant to knock your site offline), and where they go: email, webhooks (automatic messages sent to another app) or other tools. You can also silence alerts for a while, for example during planned maintenance. It helps you hear about trouble before customers do. It is free, though some alerts need certain plans.","useCases":["Get emailed when an outage starts","Alert on DDoS attacks","Post alerts to a webhook"],"pricing":"Free; some alerts need plans.","status":"ga","url":"https://developers.cloudflare.com/notifications/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-graphql-analytics-api","provider":"cloudflare","name":"GraphQL Analytics API","category":"analytics","also":["devtools"],"tagline":"Pull your Cloudflare traffic and security numbers into your own dashboards.","description":"The GraphQL Analytics API lets your code ask Cloudflare for your numbers directly, covering requests, security events and other datasets. GraphQL is a way of requesting exactly the data you want in one go. If you want a custom dashboard that shows website traffic next to your sales, you can pull the figures into it instead of reading Cloudflare's screens. Pick it when the built-in charts do not show what you need. It is included.","useCases":["Build custom dashboards with traffic numbers","Pull traffic next to sales figures","Request exact security and request data"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/analytics/graphql-api/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-dns","provider":"cloudflare","name":"Cloudflare DNS","category":"dns","tagline":"Free, fast DNS that points your domain name at your app.","description":"Cloudflare DNS is the phonebook for your domain: it turns a name like yourapp.com into the address of your server and holds the official records, which is what authoritative means. When you launch on a new domain, you set up its records here so visitors land in the right place. It uses Anycast, meaning many locations around the world can answer, which keeps lookups fast. It also comes with analytics, DNSSEC (a way to prove answers are genuine) and support for using more than one DNS provider. It is free.","useCases":["Point your domain at your app","Set up records for a new launch","Keep lookups fast worldwide"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/dns/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-dnssec","provider":"cloudflare","name":"DNSSEC and multi-provider DNS","category":"dns","tagline":"Lock down your domain's DNS and run it with a second provider.","description":"DNSSEC adds a digital signature to your domain's DNS records, the entries that point your name at your app, so browsers can tell the answers are genuine and not forged. Cloudflare turns it on with one click. Multi-signer and multi-provider DNS let you run Cloudflare alongside another DNS provider, with both signing your records. If your app cannot afford downtime, the second provider is a backup when one has an outage. It is free.","useCases":["Sign DNS records against forged answers","Run a second DNS provider as backup","Keep your domain up during outages"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/dns/dnssec/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-dns-firewall","provider":"cloudflare","name":"DNS Firewall","category":"dns","also":["ddos"],"tagline":"Shield your own DNS servers from attacks and make them answer faster.","description":"DNS Firewall puts a protective layer in front of your own nameservers, the servers that answer the question \"where does this domain live?\" If you run your own DNS and worry that a flood of junk traffic could knock your site offline, this takes the hit for you and can speed up the answers. Pick it over moving everything to Cloudflare when you want to keep your existing nameservers but still add defense. It is an Enterprise product.","useCases":["Protect your nameservers from floods","Speed up DNS answers for visitors","Add defense to existing nameservers"],"pricing":"Enterprise.","status":"ga","url":"https://developers.cloudflare.com/dns/dns-firewall/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-internal-dns","provider":"cloudflare","name":"Internal DNS","category":"dns","also":["networking"],"tagline":"Friendly names for your private tools, visible only to your team.","description":"Internal DNS lets private names, like a hostname only your team uses, point to the right place inside your company network. If your team reaches internal tools through Cloudflare One, Cloudflare's service for secure access to private resources, you can give those tools readable names instead of making people memorize addresses. Pick it when your names should work for your own people and not the public internet. It comes with Cloudflare One.","useCases":["Give private tools readable names","Name internal hosts for your team","Reach Cloudflare One tools by name"],"pricing":"Cloudflare One.","status":"ga","url":"https://developers.cloudflare.com/dns/internal-dns/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-1111-resolver","provider":"cloudflare","name":"1.1.1.1 resolver","category":"dns","tagline":"A free, privacy-focused way for devices to look up website addresses.","description":"1.1.1.1 is a free public service that looks up website addresses for you, a job your internet provider normally does. Set it on your laptop, office network or home router if you want fast lookups with a focus on privacy. It supports DoH and DoT, two ways of encrypting those lookups so others on the network can't easily see them. Pick it over your provider's default when you want a free, privacy-minded alternative.","useCases":["Set a free, private lookup service","Encrypt lookups with DoH or DoT","Speed up address lookups on your network"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/1.1.1.1/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-registrar","provider":"cloudflare","name":"Cloudflare Registrar","category":"domains","tagline":"Buy and renew domain names at cost, with no markup.","description":"Cloudflare Registrar is where you buy and renew your domain name, the web address like yourbrand.com. Prices are at wholesale cost, so you aren't paying a markup on every renewal. It includes free DNSSEC, which helps stop forged answers about your domain, and WHOIS privacy, which keeps your personal contact details out of public lookups. If you're launching a product and want a low-cost, no-fuss home for your domain, or want to transfer one in, this is a natural pick.","useCases":["Buy and renew domain names at cost","Transfer an existing domain in","Hide contact details from public lookups"],"pricing":"At-cost wholesale pricing.","status":"ga","url":"https://developers.cloudflare.com/registrar/","source":"https://developers.cloudflare.com/products/","gem":"Domains at wholesale cost, with no renewal markup."},{"id":"cloudflare-ssl-tls","provider":"cloudflare","name":"SSL/TLS edge certificates","category":"certificates","tagline":"Free HTTPS padlock for your site, issued and renewed for you.","description":"These certificates give your site the padlock and the https:// address in browsers, and they cost nothing. Cloudflare issues and renews them automatically, so you never have to remember an expiry date. You can also choose an encryption mode, which decides how strictly the connection to your own server is protected. If you're putting a new site live and want HTTPS without buying or managing certificates, this is included on every plan.","useCases":["Get a free HTTPS padlock","Renew certificates without manual reminders","Skip buying certificates at launch"],"pricing":"Free on all plans.","status":"ga","url":"https://developers.cloudflare.com/ssl/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-advanced-certificate-manager","provider":"cloudflare","name":"Advanced Certificate Manager","category":"certificates","tagline":"More control over your HTTPS certificates, as a paid add-on.","description":"Advanced Certificate Manager is a paid add-on for when the free HTTPS certificate isn't flexible enough. You can choose the certificate authority (the company that issues the certificate), how long it lasts and which hostnames it covers. It also supports custom hostnames and wildcard coverage, meaning one certificate for every subdomain, like app.yourbrand.com and blog.yourbrand.com. Pick it if a customer or security team has specific rules; otherwise the free certificate is usually enough.","useCases":["Pick the certificate authority and lifespan","Cover all subdomains with one certificate","Meet specific customer or security rules"],"pricing":"Paid add-on.","status":"ga","url":"https://developers.cloudflare.com/ssl/edge-certificates/advanced-certificate-manager/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloudflare-for-saas","provider":"cloudflare","name":"Cloudflare for SaaS (Custom Hostnames)","category":"certificates","also":["cdn"],"tagline":"Let customers use their own domain with your product, securely.","description":"If your product lets customers use their own web address, like app.theirbrand.com instead of yourproduct.com/theirbrand, this service handles it. Cloudflare issues the certificates and routes traffic for each customer-owned domain, and those domains get Cloudflare's security and performance. Pick it over building certificate handling yourself when you expect many customers with vanity domains. There is a free tier, then you pay per hostname.","useCases":["Let customers use their own domains","Issue certificates for customer domains","Route traffic for vanity domains"],"pricing":"Free tier, then per hostname.","status":"ga","url":"https://developers.cloudflare.com/cloudflare-for-platforms/cloudflare-for-saas/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-origin-ca","provider":"cloudflare","name":"Origin CA certificates","category":"certificates","tagline":"Free certificates that secure the link from Cloudflare to your server.","description":"Origin CA certificates secure the hop between Cloudflare and your own server, often called the origin. They are free and long-lived, and they are trusted by Cloudflare, so you skip buying a certificate just for that connection. They are what you need for Full (strict) mode, the tightest encryption setting. Authenticated Origin Pulls is included too, which lets your server confirm that requests really come from Cloudflare. Pick it when your site sits behind Cloudflare.","useCases":["Secure the link to your server","Enable Full strict encryption mode","Verify requests come from Cloudflare"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/ssl/origin-configuration/origin-ca/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-mtls-client-certificates","provider":"cloudflare","name":"mTLS client certificates","category":"certificates","also":["zero-trust"],"tagline":"Let devices and apps prove who they are with a certificate.","description":"mTLS client certificates let a device or app prove its identity, instead of only the server doing so. Cloudflare can issue them, or you can bring your own certificate authority, the system that creates certificates. If you have an API that only your mobile app or partner servers should call, or company devices that should reach internal tools, a certificate works like an ID card instead of a password anyone can share. Availability depends on your plan.","useCases":["Identify devices with certificates instead of passwords","Limit API access to your mobile app","Let partner servers prove their identity"],"pricing":"Plan dependent.","status":"ga","url":"https://developers.cloudflare.com/ssl/client-certificates/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-keyless-ssl","provider":"cloudflare","name":"Keyless SSL","category":"certificates","also":["secrets"],"tagline":"Use Cloudflare's HTTPS while your secret keys stay with you.","description":"Keyless SSL lets you use Cloudflare's HTTPS protection while your private keys, the secret files that prove your site is really yours, stay in your own infrastructure or HSM (a locked hardware device built to store keys). It matters when a regulator or a customer such as a bank says you can't hand your keys to anyone else. Pick it over standard certificates only if compliance requires it. It is an Enterprise product.","useCases":["Keep private keys with your own team","Meet rules that forbid handing keys over","Use HTTPS while keys stay local"],"pricing":"Enterprise.","status":"ga","url":"https://developers.cloudflare.com/ssl/keyless-ssl/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-geo-key-manager","provider":"cloudflare","name":"Geo Key Manager","category":"certificates","also":["secrets"],"tagline":"Choose which regions of the world your HTTPS keys can be stored in.","description":"Geo Key Manager lets you restrict where your private keys, the secret files behind your site's HTTPS, are stored to regions you choose. If you sell to customers or industries with data residency rules, which say certain data must stay in certain countries or regions, this helps you meet them. Pick it when a contract or regulation names where your keys may live. It is an Enterprise product.","useCases":["Restrict regions where keys are stored","Meet data residency rules for customers","Satisfy contracts that name key locations"],"pricing":"Enterprise.","status":"ga","url":"https://developers.cloudflare.com/ssl/edge-certificates/geokey-manager/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-data-localization-suite","provider":"cloudflare","name":"Data Localization Suite","category":"management","tagline":"Control which regions Cloudflare stores and inspects your data in.","description":"The Data Localization Suite is a bundle of tools for controlling where Cloudflare stores and inspects your data. It includes Regional Services, Customer Metadata Boundary and Geo Key Manager. If you sell to customers in places with strict data residency rules, which require data to stay in certain regions, this gives you one package to address that. Pick it when legal or compliance teams ask where your traffic and data are handled. It is an Enterprise product.","useCases":["Control regions where data is stored","Answer legal questions about data location","Bundle residency tools in one package"],"pricing":"Enterprise.","status":"ga","url":"https://developers.cloudflare.com/data-localization/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-ai-crawl-control","provider":"cloudflare","name":"AI Crawl Control","category":"bots","also":["analytics"],"tagline":"See which AI bots visit your site and decide which ones get in.","description":"AI Crawl Control, formerly called AI Audit, shows you which AI crawlers (automated programs that read websites to collect content) are visiting your site. If you publish articles or a blog and don't want them used to train AI models, you can see who is visiting and set rules to allow or block each crawler. It also shows whether bots actually follow your robots.txt file, the public note that tells bots what they may read. It is available on all plans, so you can try it without upgrading.","useCases":["See which AI bots visit","Block AI scrapers from your content","Check whether bots obey robots.txt"],"pricing":"Available on all plans.","status":"ga","url":"https://developers.cloudflare.com/ai-crawl-control/","source":"https://developers.cloudflare.com/ai-crawl-control/"},{"id":"cloudflare-pay-per-crawl","provider":"cloudflare","name":"Pay per crawl","category":"bots","also":["specialized"],"tagline":"Charge AI crawlers a fee each time they read your content.","description":"Pay per crawl lets AI crawlers, the automated programs that read websites to collect content, pay to access your pages. If you run a newsletter, a blog or a research site, you could earn money from AI companies that want your writing instead of only blocking them. It is the one to look at when you want a paid option rather than a plain yes or no. It is in private beta, so access is limited and no public pricing is listed.","useCases":["Charge AI crawlers for each page read","Earn money from AI content requests","Offer paid access instead of blocking"],"pricing":"Private beta.","status":"beta","url":"https://developers.cloudflare.com/ai-crawl-control/","source":"https://developers.cloudflare.com/ai-crawl-control/","gem":"Charge AI crawlers for reading your site."},{"id":"cloudflare-agent-readiness","provider":"cloudflare","name":"Agent Readiness","category":"analytics","also":["agents"],"tagline":"Get a score for how easily AI agents can find and use your site.","description":"Agent Readiness checks how prepared your website is for AI agents, which are programs that browse and act on the web for people. If you are launching a product site and want AI assistants to find and understand it, run the scan to see how you do on discoverability, content, bot access control and capabilities. You can use it in the Cloudflare dashboard or at isitagentready.com. The scan is free, which makes it a quick first check before you spend time fixing gaps.","useCases":["Score how ready your site is","Check AI agent discoverability before launch","Find gaps to fix in standards"],"pricing":"Free scan.","status":"ga","url":"https://isitagentready.com/","source":"https://blog.cloudflare.com/agent-readiness"},{"id":"cloudflare-webmcp","provider":"cloudflare","name":"WebMCP","category":"agents","tagline":"Make your site usable by visitors' AI agents with one switch.","description":"WebMCP lets the AI agents your visitors bring along use tools on your site. Cloudflare adds a small bridge that registers those tools for each agent, so you don't have to change anything on your own server. If you run a booking or store site and want visitors' browser assistants to do things there instead of guessing at buttons, this is a quick way to try it. It is a developer preview inside Agent Readiness, so expect changes along the way.","useCases":["Let visitors' AI agents use site tools","Turn on agent tools without server changes","Let visitors' assistants act on your site"],"pricing":"Developer preview.","status":"preview","url":"https://developers.cloudflare.com/browser-run/features/webmcp/","source":"https://blog.cloudflare.com/webmcp/"},{"id":"cloudflare-monetization-gateway","provider":"cloudflare","name":"Monetization Gateway","category":"specialized","also":["agents"],"tagline":"Charge people and AI agents to use your APIs, tools, pages or data.","description":"Monetization Gateway lets you put a price on what your site offers, such as APIs (ways for other software to talk to your app), MCP tools (actions AI agents can call), pages and datasets. If you built a useful data feed, it matches incoming requests, charges the amount you set and settles payment through Coinbase's x402 Facilitator, a payment service for web requests. It fits charging per request, such as paying for each tool call. It is in closed beta, US only, and no pricing is published.","useCases":["Charge per API call or tool use","Set prices for pages and datasets","Sell access to your data feed"],"pricing":"Closed beta; no pricing published.","status":"beta","url":"https://developers.cloudflare.com/monetization-gateway/","source":"https://developers.cloudflare.com/monetization-gateway/"},{"id":"cloudflare-cloudforce-one","provider":"cloudflare","name":"Cloudforce One","category":"threat-intel","tagline":"Cloudflare's threat intelligence: attacks, warning signs and bad actors.","description":"Cloudforce One is Cloudflare's threat intelligence service, which tells you what attackers are doing across the internet. Its dashboard section offers Threat Events, Threat Signals and Indicators, which are clues such as suspicious addresses that point to an attack. If your company has someone focused on security and you want to hunt for threats or feed those clues into your own tools, this gives you the data. It is sold through sales, so you need to contact Cloudflare for pricing.","useCases":["Search for threats across the internet","Feed warning signs into your tools","Spot suspicious addresses tied to attacks"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/security-center/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-url-scanner","provider":"cloudflare","name":"URL Scanner","category":"threat-intel","tagline":"Paste a link and check whether the page is safe.","description":"URL Scanner analyzes a web page and helps you see whether it looks dangerous. If someone on your team gets a suspicious email or a customer reports a strange link, you can scan it from the Investigate section instead of opening it yourself. This is handy for sorting out phishing, where fake pages trick people into handing over passwords. It is included with Cloudflare, so there is no separate fee to check a link.","useCases":["Check suspicious links without opening them yourself","Sort out phishing emails and fake pages","Scan links customers report as strange"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/security-center/investigate/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-security-center","provider":"cloudflare","name":"Security Insights (Security Center)","category":"threat-intel","also":["management"],"tagline":"Spot security risks in your account and get one-click fixes.","description":"Security Insights, also called Security Center, is a checklist of security risks across what you have set up in Cloudflare. If you launched a site quickly and aren't sure it is locked down, it highlights problems and recommends what to do next, with one-click fixes. Use it for a regular posture review, meaning a check on how well protected you are, before you reach for specialist security tools. It is included, so there is nothing extra to buy.","useCases":["Review your account for security risks","Get one-click fixes for flagged problems","Check a new site is locked down"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/security-center/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-radar","provider":"cloudflare","name":"Cloudflare Radar","category":"analytics","also":["threat-intel"],"tagline":"Free public data on internet traffic, outages, attacks and bots.","description":"Cloudflare Radar is a free public website of data about how the internet is behaving, including usage, outages and bots. If your app suddenly gets slow or users in one country can't reach it, you can check Radar to see whether a wider outage is happening. It is also useful when you want to write about trends, like how much web traffic comes from bots. Unlike tools tied to your own account, it shows the whole internet, and it costs nothing.","useCases":["Check whether an outage is widespread","Research trends in bot and web traffic","Look up internet traffic data for free"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/radar/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-web3-gateways","provider":"cloudflare","name":"Web3 gateways","category":"specialized","tagline":"Ethereum and IPFS gateways, closed to new sign-ups.","description":"Web3 gateways let an app connect to blockchain networks, namely Ethereum (a public blockchain) and IPFS (a way to store files across many computers), without running its own servers. If you already have an older app that reads from these networks through Cloudflare, this is the piece it relies on. It is mostly of interest for legacy setups. The service is closed to new sign-ups, so for a new project you will need to pick another option.","useCases":["Keep an older app reading blockchain data","Connect to Ethereum and IPFS networks","Maintain an existing app on Ethereum"],"pricing":"Closed.","status":"closed","url":"https://developers.cloudflare.com/web3/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-time-services","provider":"cloudflare","name":"Time Services","category":"specialized","also":["networking"],"tagline":"Free, public clocks your servers can use to keep accurate time.","description":"Time Services are free public clocks run by Cloudflare, offered through NTP (the standard way computers sync their clocks), NTS (a secured version of NTP) and Roughtime (another way to check the time securely). If you run your own servers and need their clocks to agree, for things like logs or login codes that expire, you can point them at Cloudflare. Pick it when you want a trusted time source without paying for one. It is free.","useCases":["Keep server clocks in agreement","Time logs and expiring login codes accurately","Use a free, trusted time source"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/time-services/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-randomness-beacon","provider":"cloudflare","name":"Randomness Beacon","category":"specialized","tagline":"A free public source of random numbers anyone can verify.","description":"Randomness Beacon is a free public source of random numbers that anyone can check was not rigged. It is built on drand, a network of independent servers that produce randomness together. If you run a giveaway or a lottery and want participants to trust how the winner was picked, you can base the draw on the beacon's numbers and show people how to verify them. Choose it over your own random number generator when fairness has to be provable to outsiders.","useCases":["Run giveaways with a provably fair winner","Pick lottery winners anyone can verify","Let outsiders check how a draw was made"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/randomness-beacon/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-ohttp-relay","provider":"cloudflare","name":"OHTTP Relay","category":"networking","also":["identity"],"tagline":"Send requests without the receiving service learning who you are.","description":"OHTTP Relay is a go-between that passes along your users' encrypted requests, so the service receiving them never learns who sent them. It follows Oblivious HTTP, a standard from the IETF (the group that sets internet standards), for private requests. If your app sends sensitive lookups or usage data and you don't want them tied to a person's IP address (the number that identifies a device online), you can route them through a relay. Pick it over a plain proxy when you want privacy by design, since the relay can't read the content and the destination can't see the sender.","useCases":["Send sensitive lookups without exposing the sender","Keep usage data unlinked from user IP addresses","Route private requests through a relay"],"pricing":"Not stated.","status":"ga","url":"https://developers.cloudflare.com/ohttp-relay/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-privacy-proxy","provider":"cloudflare","name":"Privacy Proxy","category":"networking","tagline":"Hide your users' IP addresses while keeping their rough location.","description":"Privacy Proxy hides your users' IP addresses (the number that identifies a device online) while still letting websites know roughly where they are. It uses MASQUE, a modern standard for tunneling traffic through a proxy. If you build a browser, app or service where people want private browsing, but local content, local pricing or region rules still need to work, this fits. Pick it over a typical VPN or proxy when losing location accuracy would break things for your users.","useCases":["Hide users' IP addresses while browsing","Keep local content and pricing working","Offer private browsing with region rules intact"],"pricing":"Not stated.","status":"ga","url":"https://developers.cloudflare.com/privacy-proxy/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-key-transparency","provider":"cloudflare","name":"Key Transparency Auditor","category":"identity","also":["secrets"],"tagline":"Check that a messaging app's encryption keys haven't been tampered with.","description":"Key Transparency Auditor checks the public list of encryption keys that a secure messaging service hands out, to confirm it hasn't been altered. In end-to-end encrypted (E2EE) messaging, only the sender and receiver can read messages, but people still have to trust they received the right key for the right person. If you run a chat or messaging product with E2EE, an independent auditor helps show your key directory is honest. Pick it when you want outside verification rather than asking users to just trust you.","useCases":["Verify a chat app's encryption keys","Show outside auditors your key directory is honest","Prove users get the right encryption key"],"pricing":"Not stated.","status":"ga","url":"https://developers.cloudflare.com/key-transparency/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-multi-cloud-networking","provider":"cloudflare","name":"Multi-Cloud Networking","category":"networking","tagline":"Find your resources across clouds and connect them automatically.","description":"Multi-Cloud Networking finds the resources you have running in different public clouds and builds the connections between them for you. If your database lives in one cloud and your app servers in another, you would normally set up those links by hand. Pick it when your setup is spread across more than one cloud and you want the connections created automatically instead of hand-built. Pricing is handled through Cloudflare's sales team.","useCases":["Connect resources running in different clouds","Link a database in one cloud to apps elsewhere","Find resources spread across several clouds"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/multi-cloud-networking/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-terraform-provider","provider":"cloudflare","name":"Terraform and Pulumi providers","category":"devtools","tagline":"Set up and manage Cloudflare by writing code instead of clicking.","description":"These plug-ins let you describe your Cloudflare setup in files using Terraform or Pulumi, two popular tools for infrastructure as code (managing cloud setup through files instead of clicking in a dashboard). If you want your settings to be identical every time, or you need to recreate them for a new project, you write them once and apply them whenever you like. Pick this over the dashboard when you want your setup saved, reviewable and repeatable. It's free.","useCases":["Save your Cloudflare setup as code","Recreate the same setup for new projects","Review and repeat infrastructure changes"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/terraform/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-cloudflare-cli","provider":"cloudflare","name":"Cloudflare CLI (cf)","category":"devtools","tagline":"One command-line tool to manage your Cloudflare resources.","description":"The Cloudflare CLI, called cf, is a single tool you type commands into (a command-line interface) to manage your Cloudflare resources. If you want to automate routine tasks, like running the same steps in a script every time you launch a project, you can do it without opening the dashboard. Pick it when you prefer quick typed commands and want one tool for everything. It's free.","useCases":["Automate routine tasks from scripts","Manage Cloudflare with typed commands","Skip the dashboard for quick changes"],"pricing":"Free.","status":"ga","url":"https://developers.cloudflare.com/cf/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-tenant","provider":"cloudflare","name":"Cloudflare Tenant","category":"management","tagline":"Create and manage Cloudflare accounts for your team or customers.","description":"Cloudflare Tenant lets you create and manage Cloudflare accounts and services on behalf of others, whether that's teams inside your company or your own customers. If you run a platform or agency where every customer needs their own setup, you can provision it for them instead of asking each one to sign up. It's built with resellers in mind. Pick it when you need to hand out many accounts in an organized way. Pricing is through sales.","useCases":["Create accounts for each of your customers","Set up team accounts in one organized way","Provision Cloudflare for clients as a reseller"],"pricing":"Contact sales.","status":"ga","url":"https://developers.cloudflare.com/tenant/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-resource-tagging","provider":"cloudflare","name":"Resource Tagging","category":"management","tagline":"Add labels to your resources to organize them and track costs.","description":"Resource Tagging lets you attach short labels, written as a key and a value like project: blog, to the things you run on Cloudflare. If you want to know which project or customer is driving your bill, tags let you group resources and attribute costs. They can also help with organizing things and controlling who can access what. Pick it once you have more than a handful of resources to keep track of. It's included.","useCases":["Tag resources by project or customer","Work out which project drives costs","Group and organize resources with labels"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/resource-tagging/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-audit-logs","provider":"cloudflare","name":"Audit logs","category":"management","tagline":"See who changed what in your Cloudflare account.","description":"Audit logs keep a record of changes made in your account, so you can see who changed what. If something suddenly breaks, or a teammate asks why a setting is different, you can look back through the record instead of guessing. They also help with compliance, meaning proving to customers or regulators that you keep track of changes. Pick it as your safety net whenever more than one person has access. It's included.","useCases":["See who changed a setting","Look back when something suddenly breaks","Show customers you track account changes"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/fundamentals/account/account-security/audit-logs/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-version-management","provider":"cloudflare","name":"Version Management","category":"devtools","also":["management"],"tagline":"Test settings changes first, then roll back if something goes wrong.","description":"Version Management gives your Cloudflare settings a version history, a staging step and the ability to roll back. If you're about to change a rule that affects live traffic, you can try it out first and undo it if it goes wrong. It makes updates less risky for teams where one mistake could affect real visitors. Pick it when changes need care rather than a quick edit in the dashboard. It's offered on Enterprise plans.","useCases":["Test settings changes before going live","Roll back a rule that goes wrong","Keep a version history of settings"],"pricing":"Enterprise.","status":"ga","url":"https://developers.cloudflare.com/version-management/","source":"https://developers.cloudflare.com/products/"},{"id":"cloudflare-carbon-impact-report","provider":"cloudflare","name":"Carbon Impact Report","category":"management","tagline":"See an estimate of the emissions you avoided by using Cloudflare.","description":"Carbon Impact Report is an account report that estimates how much carbon emissions you avoided by using Cloudflare. If a customer, investor or partner asks about your environmental footprint, you can pull a number from the report instead of researching it yourself. It's useful for ESG reporting, which means sharing environmental, social and governance results. Pick it when you need a quick figure for sustainability reporting. It's included.","useCases":["Get an emissions estimate for investors","Answer customer questions about your footprint","Pull a figure for sustainability reporting"],"pricing":"Included.","status":"ga","url":"https://developers.cloudflare.com/fundamentals/","source":"https://developers.cloudflare.com/products/"},{"id":"aws-ec2","provider":"aws","name":"Amazon Elastic Compute Cloud (EC2)","category":"vms","tagline":"Rent virtual servers in the cloud and resize them as you need.","description":"Amazon Elastic Compute Cloud (EC2) is Amazon's service for renting virtual servers, which are computers you use over the internet. You choose the type, the operating system and the storage, and pay for the time it runs. If you're hosting a web app, running a big batch job, or training an AI model on GPUs (chips built for heavy math), EC2 is where you'd do it. Pick it over simpler hosting when you need control or specialized hardware, since it offers the widest range of processor, GPU and custom chip options of any cloud.","useCases":["Host web and app servers yourself","Run large batch jobs on demand","Train AI models on GPU hardware"],"pricing":"Per-second on-demand, Savings Plans and Spot discounts. Free tier for new accounts.","status":"ga","url":"https://aws.amazon.com/ec2/","source":"https://aws.amazon.com/products/"},{"id":"aws-lightsail","provider":"aws","name":"Amazon Lightsail","category":"vms","also":["hosting"],"tagline":"Simple servers with a flat monthly price, starting around $5.","description":"Amazon Lightsail gives you a small server, storage, a fixed address and a data allowance bundled into one flat monthly price. If you are launching a WordPress site, a side project or a small app and want a bill that never surprises you, this is a friendly place to start. It also offers managed databases, load balancers, containers and a content delivery network (a service that serves your site faster to visitors around the world). Pick it over the full EC2 server service when you want fewer settings and predictable costs. New plans come with a free trial.","useCases":["Launch a WordPress site on a fixed plan","Run dev and test servers cheaply","Host a small app with a fixed bill"],"pricing":"Fixed monthly plans from about $5. Free trial for new plans.","status":"ga","url":"https://aws.amazon.com/lightsail/","source":"https://aws.amazon.com/products/"},{"id":"aws-lambda","provider":"aws","name":"AWS Lambda","category":"serverless","tagline":"Run small pieces of code on demand and pay only when they run.","description":"AWS Lambda runs a piece of your code whenever something happens, like a web request, a message in a queue or a file upload, with no server for you to look after. If your app lets people upload photos and you want each one resized automatically, Lambda can do that job. AWS handles scaling and security patches for you. It is the default way to run serverless code on AWS, and you pay per request and per millisecond. The free tier includes 1 million requests each month.","useCases":["Resize photos automatically after each upload","Handle web requests without managing servers","Run scheduled jobs without a server"],"pricing":"Usage-based. Free tier: 1M requests and 400,000 GB-seconds per month.","status":"ga","url":"https://aws.amazon.com/lambda/","source":"https://aws.amazon.com/products/"},{"id":"aws-fargate","provider":"aws","name":"AWS Fargate","category":"containers","also":["serverless"],"tagline":"Run your app's containers without looking after any servers.","description":"AWS Fargate runs containers, which are packaged-up apps that run the same way everywhere, without making you manage the servers underneath. If you have a web service or a background worker already packaged as a container, you tell Fargate how much CPU and memory it needs and it runs it. You pay only for what each task uses while it runs. Choose it over running your own servers when you want containers without the upkeep. It works with both ECS and EKS, the two main AWS container services.","useCases":["Run containerized web services without servers","Run background workers as containers","Run batch tasks, paying only while active"],"pricing":"Per vCPU and GB-second while tasks run.","status":"ga","url":"https://aws.amazon.com/fargate/","source":"https://aws.amazon.com/products/"},{"id":"aws-batch","provider":"aws","name":"AWS Batch","category":"workflows","also":["containers"],"tagline":"Queue up big batches of jobs and let AWS find the computers to run them.","description":"AWS Batch lines up large numbers of jobs, then starts the right amount of computing power to run them and shuts it down afterward. If you need to process thousands of video files or prepare a big pile of data for machine learning, you hand Batch the list and it does the scheduling. That saves you from building your own job queue and managing a group of servers. Batch itself costs nothing, and you pay only for the computing it uses. It can use regular servers, cheaper spare-capacity Spot servers or Fargate.","useCases":["Process thousands of video files","Run large queues of jobs","Prepare big datasets for machine learning"],"pricing":"No charge for Batch itself. Pay for the compute used.","status":"ga","url":"https://aws.amazon.com/batch/","source":"https://aws.amazon.com/products/"},{"id":"aws-elastic-beanstalk","provider":"aws","name":"AWS Elastic Beanstalk","category":"hosting","tagline":"Upload your web app and AWS sets up everything it needs to run.","description":"AWS Elastic Beanstalk takes your web app code and sets up the hosting around it, including traffic balancing, scaling and monitoring. If you built a Node.js, Python, PHP, Ruby, Java, .NET or Go app and want it online without learning server setup, you upload it and Beanstalk does the rest. Unlike some simpler hosts, you can still reach and adjust the underlying resources if you need to. Beanstalk is free, and you pay only for the resources it creates for your app.","useCases":["Upload a web app and get it online","Host simple apps without learning server setup","Spin up dev and staging environments"],"pricing":"No charge for Beanstalk. Pay for the resources it creates.","status":"ga","url":"https://aws.amazon.com/elasticbeanstalk/","source":"https://aws.amazon.com/products/"},{"id":"aws-app-runner","provider":"aws","name":"AWS App Runner","category":"hosting","also":["containers"],"tagline":"Run a web app or API straight from your code or container.","description":"AWS App Runner builds and runs a web app or API directly from a code repository or a container image, which is a packaged-up version of your app. It handles scaling, traffic balancing and secure HTTPS connections automatically. If you have a prototype you want online quickly, it was a low-fuss way to get there. You paid per CPU and memory hour while the app was active. App Runner closed to new customers on April 30, 2026, and ECS Express Mode is suggested as the replacement.","useCases":["Run a web app straight from code","Put a prototype online quickly","Serve a simple API from a container"],"pricing":"Per vCPU and GB-hour for active instances.","status":"closed","url":"https://docs.aws.amazon.com/apprunner/latest/dg/what-is-apprunner.html","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-image-builder","provider":"aws","name":"EC2 Image Builder","category":"devtools","also":["vms"],"tagline":"Automatically build and update ready-to-use server images.","description":"EC2 Image Builder creates server images, which are saved copies of a server set up exactly how you want it, and keeps them up to date. If your team launches many servers and each needs the same security patches and software, you set up a pipeline that builds a fresh, tested image on a schedule. It also makes container base images. Use it instead of updating images by hand, which is slow and easy to get wrong. The tool is free, and you pay only for the resources it uses.","useCases":["Build server images with security updates included","Rebuild updated images on a schedule","Create base images for containers"],"pricing":"No charge for Image Builder. Pay for the underlying resources.","status":"ga","url":"https://aws.amazon.com/image-builder/","source":"https://aws.amazon.com/products/"},{"id":"aws-outposts","provider":"aws","name":"AWS Outposts","category":"vms","also":["networking"],"tagline":"AWS servers and services installed in your own building.","description":"AWS Outposts puts AWS-designed hardware in your own data center or facility, running familiar AWS services like EC2 servers, EBS storage and RDS databases through the same tools as the cloud. If you run a factory where machines need answers in a split second, or you must keep data in a specific location, Outposts keeps the computing close by. Choose it when regular cloud regions are too far away or not allowed. Pricing is a fixed term per rack or server, so you contact sales to get a quote.","useCases":["Run AWS services inside your own building","Keep data in a required location","Run factory systems that need instant responses"],"pricing":"Contact sales. Fixed term capacity pricing per rack or server.","status":"ga","url":"https://aws.amazon.com/outposts/","source":"https://aws.amazon.com/products/"},{"id":"aws-local-zones","provider":"aws","name":"AWS Local Zones","category":"vms","also":["networking"],"tagline":"AWS computing placed in big cities, closer to your users.","description":"AWS Local Zones extends an AWS region into major metro areas so your app runs physically closer to the people using it. If you run a multiplayer game or a live video production tool where delay is noticeable, a nearby Local Zone cuts that lag. You use the same console and tools as usual, though fewer AWS services are available. Pick it over a standard region when speed for a specific city matters most. Pricing follows the parent region, with Local Zone rates for each resource.","useCases":["Run live video tools close to city users","Host multiplayer games with less lag","Power real-time dashboards for a specific city"],"pricing":"Same model as the parent Region, with Local Zone pricing per resource.","status":"ga","url":"https://aws.amazon.com/about-aws/global-infrastructure/localzones/","source":"https://aws.amazon.com/products/"},{"id":"aws-wavelength","provider":"aws","name":"AWS Wavelength","category":"vms","also":["edge-compute"],"tagline":"AWS computing built into 5G phone networks for ultra-fast response.","description":"AWS Wavelength places AWS computing and storage inside phone carriers' data centers at the edge of 5G networks. If you are building an augmented reality app, live game streaming or connected vehicle features, your app can reach mobile devices with response times in the single-digit milliseconds. That is faster than sending traffic to a distant cloud region. Choose it when mobile speed is the whole point of your product. You pay usage-based EC2 server prices in Wavelength Zones.","useCases":["Run augmented reality apps for phone users","Stream live games to mobile devices","Support features in connected vehicles"],"pricing":"Usage-based. EC2 pricing in Wavelength Zones.","status":"ga","url":"https://aws.amazon.com/wavelength/","source":"https://aws.amazon.com/products/"},{"id":"aws-parallel-computing-service","provider":"aws","name":"AWS Parallel Computing Service","category":"specialized","also":["vms"],"tagline":"Managed supercomputer-style clusters for heavy science and engineering.","description":"AWS Parallel Computing Service sets up and runs HPC clusters (groups of computers working together on huge calculations) using Slurm, a popular job scheduler. If your team simulates weather, airflow around a car or molecules, you can run those jobs without managing the scheduler yourself. It is the hands-off choice compared with building clusters on your own. You pay for the controller and the computing resources your jobs use.","useCases":["Run weather and airflow simulations","Model molecules without managing schedulers","Process scientific research calculations"],"pricing":"Pay for the controller and compute resources used.","status":"ga","url":"https://aws.amazon.com/pcs/","source":"https://aws.amazon.com/products/"},{"id":"aws-parallelcluster","provider":"aws","name":"AWS ParallelCluster","category":"specialized","also":["vms"],"tagline":"Free open source tool to build your own supercomputer-style clusters.","description":"AWS ParallelCluster is a free command line tool that builds HPC clusters (groups of computers working together on huge calculations) from a single settings file, using Slurm, a popular job scheduler, and EC2 servers. If your research group wants to run simulations and needs to tune every detail, you describe the cluster and ParallelCluster creates it. It gives you more control than the managed AWS Parallel Computing Service, though you run more of it yourself. The tool costs nothing, and you pay only for the resources used.","useCases":["Build your own calculation clusters for research","Set up clusters from one settings file","Add extra cloud power for peak workloads"],"pricing":"No charge for the tool. Pay for resources used.","status":"ga","url":"https://aws.amazon.com/hpc/parallelcluster/","source":"https://aws.amazon.com/products/"},{"id":"aws-res","provider":"aws","name":"Research and Engineering Studio on AWS","category":"specialized","also":["end-user"],"tagline":"Give researchers secure remote desktops and shared tools in your AWS account.","description":"Research and Engineering Studio on AWS is a free, open source web portal that lets administrators give researchers and engineers secure virtual desktops (computers you use remotely) and shared tools. If you run a lab or engineering team that needs heavy computing, including HPC (high-performance computing, meaning very large calculations), you can set up one place where people log in and work together. It runs in your own AWS account, so you stay in charge of it. It costs nothing to deploy, and you pay only for the AWS resources it uses.","useCases":["Give researchers secure remote desktops","Offer one login portal for engineering tools","Share heavy computing access across a lab"],"pricing":"Free to deploy. Pay for the AWS resources used.","status":"ga","url":"https://aws.amazon.com/hpc/res/","source":"https://aws.amazon.com/products/"},{"id":"aws-sam","provider":"aws","name":"AWS Serverless Application Model (SAM)","category":"devtools","also":["serverless"],"tagline":"Shorthand templates and a command line tool for building serverless apps.","description":"AWS SAM is a free, open source toolkit that makes serverless apps shorter to describe and launch. Serverless means AWS runs the servers for you. If you are building a small backend with Lambda (code that runs only when triggered), API Gateway (the front door for web requests) and DynamoDB (a database), SAM lets you define it all in one short file. Its command line tool can build, test on your own computer and deploy. Pick it over writing raw CloudFormation, AWS's general setup system, when you only need serverless pieces. You pay only for what you deploy.","useCases":["Deploy small serverless backends quickly","Test serverless functions on your own computer","Automate deploys of serverless functions"],"pricing":"Free. Pay for resources deployed.","status":"ga","url":"https://aws.amazon.com/serverless/sam/","source":"https://aws.amazon.com/products/"},{"id":"aws-serverless-application-repository","provider":"aws","name":"AWS Serverless Application Repository","category":"serverless","tagline":"Browse and install ready-made serverless apps with a few clicks.","description":"The AWS Serverless Application Repository is a catalog of ready-made serverless apps, meaning apps where AWS runs the servers for you, shared by AWS and community publishers. If you need a common building block, you can find one and install it into your account with a few clicks instead of building it from scratch. It also works the other way, so you can publish an app of your own for others to reuse. Browsing the catalog costs nothing, and you pay only for the resources the apps you deploy end up using.","useCases":["Install ready-made serverless app building blocks","Browse a catalog of shared apps","Publish your own app for others to reuse"],"pricing":"No charge for the repository. Pay for resources deployed.","status":"ga","url":"https://aws.amazon.com/serverless/serverlessrepo/","source":"https://aws.amazon.com/products/"},{"id":"aws-linux","provider":"aws","name":"Amazon Linux","category":"vms","tagline":"The free Linux system AWS builds and updates for running your servers.","description":"Amazon Linux is the operating system, the base software a server runs on, that AWS builds and supports for its cloud. If you launch a server on EC2, AWS's rentable servers, and have no strong preference, this is the default image to start from. It is tuned for AWS and comes with security updates at no extra cost. You can also use it as a starting point for container images or local development images. Amazon Linux 2023 is the current release. The software is free, and you pay only for the infrastructure it runs on.","useCases":["Run servers on a free, maintained system","Start container images from a trusted base","Set up local development environments"],"pricing":"Free. Pay for the infrastructure it runs on.","status":"ga","url":"https://aws.amazon.com/linux/amazon-linux-2023/","source":"https://aws.amazon.com/products/"},{"id":"aws-ecs","provider":"aws","name":"Amazon Elastic Container Service (ECS)","category":"containers","tagline":"Run your Docker containers on AWS without learning Kubernetes.","description":"Amazon ECS runs and scales Docker containers, which are apps packaged with everything they need to run. If you have a web app or a few small services in containers, you describe them as tasks and services and ECS keeps them running on EC2 servers or on Fargate, where AWS handles the servers. It is simpler than Kubernetes, a popular but more complex container system, and works closely with IAM (AWS's permissions system) and load balancers (which spread visitors across copies of your app). ECS itself is free, and you pay only for the capacity you use.","useCases":["Run web apps in containers","Keep several small services running","Run batch jobs in containers"],"pricing":"No charge for ECS. Pay for EC2 or Fargate capacity.","status":"ga","url":"https://aws.amazon.com/ecs/","source":"https://aws.amazon.com/products/"},{"id":"aws-eks","provider":"aws","name":"Amazon Elastic Kubernetes Service (EKS)","category":"containers","tagline":"Run Kubernetes on AWS while AWS looks after the control center.","description":"Amazon EKS is a managed version of Kubernetes, a popular system for running many containers (apps packaged to run anywhere). AWS runs the control plane, the part that makes the decisions, across multiple zones, so you manage only your own workers. If your team already knows Kubernetes or wants apps that can move between clouds, pick EKS over ECS. It connects with IAM (permissions), VPC (your private network) and Fargate, and can also run on-premises through EKS Anywhere and Hybrid Nodes. Expect about $0.10 per cluster-hour plus the cost of worker compute.","useCases":["Run Kubernetes-based platforms for your team","Move container apps between clouds","Run machine learning on Kubernetes"],"pricing":"About $0.10 per cluster-hour plus worker compute.","status":"ga","url":"https://aws.amazon.com/eks/","source":"https://aws.amazon.com/products/"},{"id":"aws-ecr","provider":"aws","name":"Amazon Elastic Container Registry (ECR)","category":"containers","also":["devtools"],"tagline":"A private storage shelf for your app's container images.","description":"Amazon ECR is a private place to store your container images, the packaged versions of your apps. If your app runs in containers, you push each new version here and your servers pull it when they start. It scans images for security problems and can copy them to other AWS Regions. Access is controlled with IAM, AWS's permissions system, and it works closely with ECS, EKS and Lambda. You pay per GB of storage each month plus data transfer.","useCases":["Store private versions of your app","Scan container images for security problems","Copy images to other locations"],"pricing":"Per GB-month of storage plus data transfer.","status":"ga","url":"https://aws.amazon.com/ecr/","source":"https://aws.amazon.com/products/"},{"id":"aws-rosa","provider":"aws","name":"Red Hat OpenShift Service on AWS (ROSA)","category":"containers","tagline":"Red Hat OpenShift, run jointly by AWS and Red Hat, billed through AWS.","description":"ROSA is Red Hat OpenShift, a platform built on Kubernetes (a popular system for running containers) for running business apps, offered as a managed service on AWS. AWS and Red Hat operate it jointly, and you get AWS billing and support. If your company already uses OpenShift, or you are moving existing OpenShift apps to AWS, it saves you from running the platform yourself. Choose it over EKS when OpenShift specifically is what your team wants. You pay an hourly cluster fee plus compute and infrastructure.","useCases":["Run OpenShift apps on AWS","Operate enterprise container platforms","Move existing OpenShift apps to AWS"],"pricing":"Hourly cluster fee plus compute and infrastructure.","status":"ga","url":"https://aws.amazon.com/rosa/","source":"https://aws.amazon.com/products/"},{"id":"aws-app2container","provider":"aws","name":"AWS App2Container","category":"migration","also":["containers"],"tagline":"Turn existing Java and .NET apps into containers with a command line tool.","description":"AWS App2Container is a free command line tool that helps you move older Java and .NET apps into containers, which are packaged apps that run the same way anywhere. If you have an app running on an old server that nobody wants to rewrite, the tool looks at the running app, builds the container image and creates the files needed to deploy it on ECS or EKS, AWS's container services. Pick it over rebuilding by hand when you are modernizing legacy apps. You pay only for the AWS resources you use.","useCases":["Move old Java apps into containers","Containerize older .NET applications","Modernize legacy software without rewriting"],"pricing":"Free. Pay for the AWS resources used.","status":"ga","url":"https://aws.amazon.com/app2container/","source":"https://aws.amazon.com/products/"},{"id":"aws-app-mesh","provider":"aws","name":"AWS App Mesh","category":"networking","also":["containers"],"tagline":"Traffic control between your app's services. AWS is ending support.","description":"AWS App Mesh is a service mesh, a layer that controls and watches the traffic between the small services that make up an app. It is based on Envoy, a popular traffic-handling tool. You might use it to send a small share of visitors to a new version, called a canary release, before rolling it out to everyone. There is no charge for App Mesh itself, only for the compute you run. AWS has announced it will end support for App Mesh and points users to ECS Service Connect instead.","useCases":["Control traffic between app services","Test new versions on a few visitors","Roll out changes gradually"],"pricing":"No charge for App Mesh. Pay for compute.","status":"deprecated","url":"https://aws.amazon.com/app-mesh/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-vpc-lattice","provider":"aws","name":"Amazon VPC Lattice","category":"networking","also":["load-balancing"],"tagline":"Let your app's services find, trust and talk to each other securely.","description":"Amazon VPC Lattice helps the different services in your app find, trust and talk to each other, even when they live in separate private networks (VPCs) or AWS accounts. If you split your product into microservices, small apps that each do one job, you can set rules for who may call what using IAM, AWS's permissions system. Unlike a traditional mesh, it needs no sidecar, an extra helper program running beside each service. You pay per service-hour, per GB processed and per request.","useCases":["Connect microservices across separate networks","Set rules for which services can call","Let services talk without helper programs"],"pricing":"Per service-hour, per GB processed and per request.","status":"ga","url":"https://aws.amazon.com/vpc/lattice/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloud-map","provider":"aws","name":"AWS Cloud Map","category":"networking","tagline":"A phonebook where your app's parts look each other up by name.","description":"AWS Cloud Map is a registry, like a phonebook, where the parts of your app sign up under names you choose and other parts look them up. If you run services on ECS, your app can ask for a name instead of relying on addresses that keep changing. Lookups work through DNS (the internet's naming system) or an API, and health checking tells you which resources are working. You pay per registered resource and per discovery request.","useCases":["Let app parts find each other by name","Look up service addresses that change","Check which app parts are healthy"],"pricing":"Per registered resource and per discovery request.","status":"ga","url":"https://aws.amazon.com/cloud-map/","source":"https://aws.amazon.com/products/"},{"id":"aws-api-gateway","provider":"aws","name":"Amazon API Gateway","category":"networking","also":["serverless"],"tagline":"A front door for your app's API that handles security and traffic limits.","description":"Amazon API Gateway is a managed front door for your app's API, the set of web addresses other software calls to get things done. If you have backend code on AWS Lambda, which runs your code only when something asks for it, this is the usual way to give that code a public web address. It also handles traffic limits, sign-in checks, caching and usage plans, so you can offer your API as a product. It supports REST, HTTP and WebSocket (live two-way connection) APIs. You pay per million calls, with 1 million free calls a month for 12 months.","useCases":["Give serverless code a public web address","Offer your API as a product","Add live two-way connections for apps"],"pricing":"Per million calls. Free tier: 1M calls per month for 12 months.","status":"ga","url":"https://aws.amazon.com/api-gateway/","source":"https://aws.amazon.com/products/"},{"id":"aws-appsync","provider":"aws","name":"AWS AppSync","category":"serverless","also":["realtime-db"],"tagline":"Connect your app to your data and push live updates to users.","description":"AWS AppSync lets you build a GraphQL API, a flexible way for your app to ask for exactly the data it needs, without running servers yourself. If you are building a chat feature or a live dashboard on mobile or web, its real-time subscriptions push new data to screens as it changes. It can pull from DynamoDB (a database), Lambda (code that runs on demand), OpenSearch and other web sources, and it supports offline sync. Pick it over a plain API when you want live updates or need to combine several data sources behind one API. Pricing is per million queries and real-time updates.","useCases":["Build backends for mobile and web apps","Push live updates to users' screens","Combine several data sources behind one API"],"pricing":"Per million queries and real-time updates.","status":"ga","url":"https://aws.amazon.com/appsync/","source":"https://aws.amazon.com/products/"},{"id":"aws-amplify","provider":"aws","name":"AWS Amplify","category":"hosting","also":["devtools"],"tagline":"Host your web app and set up its login, data and file storage.","description":"AWS Amplify is a hosting service plus a set of tools for building full-stack web and mobile apps. If your Next.js or React site lives in a code repository, Amplify can rebuild and publish it each time you push changes, and it can create preview versions so you can check updates before going live. It can also set up a backend for sign-in, data and file storage from TypeScript code. Choose it when you want hosting and backend pieces in one place on AWS. You pay per build minute, hosting GB and requests, and there is a free tier.","useCases":["Host Next.js and React websites","Publish previews before going live","Add sign-in and file storage"],"pricing":"Pay per build minute, hosting GB and requests. Free tier available.","status":"ga","url":"https://aws.amazon.com/amplify/","source":"https://aws.amazon.com/products/"},{"id":"aws-app-studio","provider":"aws","name":"AWS App Studio","category":"devtools","also":["ai-apps"],"tagline":"Describe an internal business app in plain words and get it built.","description":"AWS App Studio builds internal business apps from a written description, using generative AI and no deep coding. If your team needs an approval workflow or a simple data entry form, you can describe it in everyday language and let the service generate the app for you. It is aimed at internal tools for your own team rather than public products for customers. It costs nothing to build, and you pay only for the underlying AWS resources your app uses.","useCases":["Build approval workflows from plain descriptions","Create internal data entry forms","Make tools for your own team"],"pricing":"Free to build. Pay for the underlying resources used.","status":"ga","url":"https://aws.amazon.com/app-studio/","source":"https://aws.amazon.com/products/"},{"id":"aws-appfabric","provider":"aws","name":"AWS AppFabric","category":"end-user","also":["observability"],"tagline":"Pull activity logs from your work apps into one tidy format for security.","description":"AWS AppFabric connects to business apps like Slack, Okta and Zoom and collects their security logs and records of who did what. It then converts them into a common format so security tools can read them all the same way. If your company uses a dozen SaaS (subscription software) tools and someone needs to watch for suspicious activity, this saves you from handling each app's different log style. It is built for security monitoring and unified audit records, not for building your own product. Pricing depends on how many apps you connect and how much data flows in.","useCases":["Watch security activity across work apps","Combine audit logs into one format","Connect Slack, Okta and Zoom logs"],"pricing":"Per connected app and ingestion volume.","status":"ga","url":"https://aws.amazon.com/appfabric/","source":"https://aws.amazon.com/products/"},{"id":"aws-transform","provider":"aws","name":"AWS Transform","category":"migration","also":["agents"],"tagline":"AI helpers that update old software and move it to the cloud.","description":"AWS Transform uses AI agents, programs that carry out tasks for you, to modernize old systems and code. If your company runs a mainframe, a .NET app (Microsoft's programming framework) or VMware (virtual machine software) servers, it helps with assessing, planning and converting the code. It can also handle upgrades, such as moving to newer Java versions. Pick it when you have existing legacy software to move rather than a new app to build. Most transformation tasks are free, while some agents are billed by usage.","useCases":["Modernize old mainframe systems","Move .NET apps to Linux","Upgrade Java code to newer versions"],"pricing":"Mostly free for transformation tasks. Usage-based for some agents.","status":"ga","url":"https://aws.amazon.com/transform/","source":"https://aws.amazon.com/products/"},{"id":"aws-transform-mgn","provider":"aws","name":"AWS Transform MGN","category":"migration","tagline":"Copy your existing servers into AWS and keep them in sync until you switch.","description":"AWS Transform MGN, formerly called Application Migration Service, moves servers you already run into Amazon EC2, the rentable virtual servers from AWS. It continuously copies each server's disk data to AWS, so you can switch over when you are ready. If you are leaving a data center or planning a disaster recovery cutover, this is the tool for the job. It now sits under the AWS Transform family name. Each server is free for 90 days, then billed hourly while it is replicating.","useCases":["Copy existing servers into AWS","Keep servers in sync until switchover","Plan disaster recovery cutovers"],"pricing":"Free for 90 days per server, then hourly per replicating server.","status":"ga","url":"https://aws.amazon.com/application-migration-service/","source":"https://aws.amazon.com/products/"},{"id":"aws-s3","provider":"aws","name":"Amazon Simple Storage Service (S3)","category":"object-storage","tagline":"Store any amount of files, images and backups online.","description":"Amazon S3 is online storage for files of any kind and any amount. If your app lets people upload photos or documents, S3 is where those files usually live. It offers storage classes (cheaper options for data you rarely touch), versioning to keep older copies, and 11 nines of durability, meaning files are extremely unlikely to be lost. It is the base layer many AWS tools build on, from backups to data lakes (big collections of raw data). You pay per GB stored each month plus requests and data sent out, with a free tier for 12 months.","useCases":["Store photos and documents users upload","Keep nightly backups of app data","Serve static files like images"],"pricing":"Per GB-month by storage class plus requests and egress. Free tier for 12 months.","status":"ga","url":"https://aws.amazon.com/s3/","source":"https://aws.amazon.com/products/"},{"id":"aws-s3-vectors","provider":"aws","name":"Amazon S3 Vectors","category":"vector","also":["object-storage"],"tagline":"Store and search the number lists AI uses to find similar content, cheaply.","description":"Amazon S3 Vectors lets you store and search vector embeddings, lists of numbers that help AI find text or images with similar meaning, right inside S3. If you are building a chatbot that answers from your own documents (often called RAG), you can keep those embeddings here instead of in a dedicated vector database. It costs less than those databases and suits large collections that you do not search very often. You pay per GB stored plus charges for adding and querying data.","useCases":["Power chatbots that answer from your documents","Search archives by meaning, not keywords","Store AI-ready search data cheaply"],"pricing":"Per GB stored plus put and query charges.","status":"ga","url":"https://aws.amazon.com/s3/features/vectors/","source":"https://aws.amazon.com/products/"},{"id":"aws-glacier","provider":"aws","name":"Amazon S3 Glacier","category":"block-storage","also":["object-storage"],"tagline":"Cheap storage for files you rarely need, kept for the long haul.","description":"Amazon S3 Glacier is a group of low-cost archive storage classes inside S3, built for data you rarely read. If you must keep old records for compliance, long-term backups or old media files, you can park them here for very little per GB each month. The trade-off is that getting files back takes minutes to hours, and retrieval comes with fees. Choose it over regular S3 when cost matters more than quick access. The older standalone Glacier vault API is in maintenance mode.","useCases":["Keep old records for compliance rules","Store long-term backups at low cost","Archive old media files cheaply"],"pricing":"Very low per GB-month plus retrieval fees.","status":"ga","url":"https://aws.amazon.com/s3/storage-classes/glacier/","source":"https://aws.amazon.com/products/"},{"id":"aws-ebs","provider":"aws","name":"Amazon Elastic Block Store (EBS)","category":"block-storage","tagline":"Hard drives for your cloud servers that keep data when the server stops.","description":"Amazon EBS provides disk drives for your EC2 servers, the rentable virtual machines from AWS. If you run a database or web app on an EC2 server, EBS holds the operating system and your data, and that data persists even when the server restarts. You can take snapshots, which are saved backup copies stored in S3, and encrypt your volumes. It is the default disk for AWS virtual machines, so you pick it unless you need shared files or object storage. Pricing is per GB each month by drive type, plus extra for added speed.","useCases":["Hold the operating system for servers","Store data for a database server","Save backup snapshots of server disks"],"pricing":"Per GB-month by volume type plus provisioned IOPS and throughput.","status":"ga","url":"https://aws.amazon.com/ebs/","source":"https://aws.amazon.com/products/"},{"id":"aws-efs","provider":"aws","name":"Amazon Elastic File System (EFS)","category":"block-storage","tagline":"A shared drive many servers can use at once that grows on its own.","description":"Amazon EFS is shared file storage that many servers can read and write at the same time, using NFS, a standard way of sharing folders over a network. If you run several containers or servers that all need the same files, such as shared content or home directories, EFS gives them one common folder. It grows and shrinks automatically as you add or remove files, so you do not plan capacity ahead. Pick it over EBS, which is built around disks for individual servers, when sharing matters. Pricing is per GB each month by storage class, plus throughput options.","useCases":["Share the same files across servers","Give containers a shared storage folder","Host shared home directories for users"],"pricing":"Per GB-month by storage class plus throughput options.","status":"ga","url":"https://aws.amazon.com/efs/","source":"https://aws.amazon.com/products/"},{"id":"aws-fsx","provider":"aws","name":"Amazon FSx","category":"block-storage","tagline":"Shared file storage for your team, run for you by AWS.","description":"Amazon FSx gives you ready-made shared file storage, like a shared drive, without you running the servers behind it. It comes in several types: Windows file shares, Lustre (built for very fast number crunching), NetApp ONTAP and OpenZFS. If your team already uses a Windows shared drive and wants it in the cloud, FSx lets people keep working the same way. Pick it over simpler storage when your software expects a traditional file system. Pricing depends on the type, based on storage, speed and backups.","useCases":["Run Windows shared drives in the cloud","Crunch large scientific data quickly","Move existing NetApp storage to AWS"],"pricing":"Per GB-month plus throughput and backups, by file system type.","status":"ga","url":"https://aws.amazon.com/fsx/","source":"https://aws.amazon.com/products/"},{"id":"aws-backup","provider":"aws","name":"AWS Backup","category":"block-storage","also":["management"],"tagline":"One place to schedule and manage backups across your AWS services.","description":"AWS Backup is a single control panel for saving copies of your data across AWS. You set rules once for how often to back up and how long to keep copies, and they apply to services like S3 (file storage), RDS and DynamoDB (databases) and more. If your product keeps customer data in several places, this saves you from setting up backups separately for each. It can also copy backups to another account or AWS location, which helps you recover from something like ransomware.","useCases":["Set one backup schedule for all services","Copy backups to a separate vault","Keep retention rules for stored copies"],"pricing":"Per GB-month of backup storage plus restore fees.","status":"ga","url":"https://aws.amazon.com/backup/","source":"https://aws.amazon.com/products/"},{"id":"aws-elastic-disaster-recovery","provider":"aws","name":"AWS Elastic Disaster Recovery","category":"block-storage","also":["migration"],"tagline":"A standby copy of your servers you can switch on if disaster hits.","description":"AWS Elastic Disaster Recovery keeps a constantly updated copy of your servers in AWS, ready to start if your main setup fails. If your app runs on servers in your own office or data center, or in a single AWS location, an outage would not mean starting from scratch. When disaster hits, it launches your servers as EC2 instances, AWS's rentable virtual servers. It was formerly called CloudEndure Disaster Recovery. Pick it when you want a recovery plan without running a full second copy of everything. You pay per hour for each server being copied, plus small staging resources.","useCases":["Keep standby copies of office servers","Switch servers on after an outage","Protect servers without a full second copy"],"pricing":"Per hour per replicating source server plus staging resources.","status":"ga","url":"https://aws.amazon.com/disaster-recovery/","source":"https://aws.amazon.com/products/"},{"id":"aws-storage-gateway","provider":"aws","name":"AWS Storage Gateway","category":"block-storage","also":["migration"],"tagline":"Connect your office storage to AWS so existing tools keep working.","description":"AWS Storage Gateway is a bridge between equipment in your own office or data center and storage in AWS. To your existing systems it looks like an ordinary file share, tape library or disk, but the data is kept in AWS, with a local cache for quick access to what you use most. If you have older software that backs up to tape or a network drive, you can point it at the gateway and store those backups in AWS instead. Pick it when you cannot move everything to the cloud yet.","useCases":["Replace old tape backups with cloud storage","Show cloud storage as a network drive","Back up office systems to AWS"],"pricing":"Per gateway plus per GB stored and transferred.","status":"ga","url":"https://aws.amazon.com/storage-gateway/","source":"https://aws.amazon.com/products/"},{"id":"aws-snowball","provider":"aws","name":"AWS Snowball Edge","category":"migration","also":["block-storage"],"tagline":"Rugged devices shipped to you for moving huge amounts of data to AWS.","description":"AWS Snowball Edge is a rugged physical device that AWS ships to you. You load your data onto it and send it back, or use it to run computing in places with poor internet. If you have petabytes (thousands of terabytes) of footage or records, shipping a box can beat uploading over the internet. It also suits remote sites that need to process data on the spot. The storage and compute optimized devices are in maintenance and closed to new customers.","useCases":["Move petabytes of records by shipping","Process data on site in remote places"],"pricing":"Per job plus shipping and per-day device fees.","status":"closed","url":"https://aws.amazon.com/snowball/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-datasync","provider":"aws","name":"AWS DataSync","category":"migration","tagline":"Copy files between your storage and AWS, with scheduling and checks.","description":"AWS DataSync is a managed tool that copies large amounts of files from one storage system to another. If your company keeps files on a network drive in the office and you want them in S3, AWS's file storage, DataSync does the move and verifies that everything arrived intact. It can also pull from other clouds and run on a schedule, so two locations stay in sync. Pick it over dragging files by hand or writing your own copy scripts when the job is big or repeating. You pay per GB copied.","useCases":["Copy office network drive files to S3","Move files between clouds","Sync two locations on a schedule"],"pricing":"Per GB copied.","status":"ga","url":"https://aws.amazon.com/datasync/","source":"https://aws.amazon.com/products/"},{"id":"aws-transfer-family","provider":"aws","name":"AWS Transfer Family","category":"migration","also":["object-storage"],"tagline":"Let partners send you files the old way, straight into AWS storage.","description":"AWS Transfer Family gives you ready-made file-transfer endpoints that speak SFTP, FTPS, FTP and AS2, older but still common ways for businesses to swap files. Behind them, files land in S3 or EFS, two AWS storage services. If a customer or supplier insists on sending you a daily file by SFTP, you can say yes without running your own server. Pick it when partners will not change their tools and you want their files in your cloud storage. You pay per hour for each protocol you turn on, plus the data transferred.","useCases":["Let partners send files by SFTP","Land partner files in S3 storage","Exchange business files with suppliers"],"pricing":"Per protocol-hour enabled plus per GB transferred.","status":"ga","url":"https://aws.amazon.com/aws-transfer-family/","source":"https://aws.amazon.com/products/"},{"id":"aws-data-transfer-terminal","provider":"aws","name":"AWS Data Transfer Terminal","category":"migration","tagline":"Walk-in sites where you plug in drives and upload data to AWS fast.","description":"AWS Data Transfer Terminal is a staffed physical location you visit to upload big datasets to AWS. You bring your storage devices, plug them into a very fast connection to AWS, and the data goes straight up. If you have a huge batch of video footage or genomics (DNA sequencing) data that would take forever over your office internet, this is a quick way to get it in. Pick it for large one-off uploads. You pay per hour of terminal use plus the transfer.","useCases":["Upload huge datasets at a visited site","Upload genomics DNA sequencing data","Skip slow office internet for big jobs"],"pricing":"Per hour of terminal use plus transfer.","status":"ga","url":"https://aws.amazon.com/data-transfer-terminal/","source":"https://aws.amazon.com/products/"},{"id":"aws-rds","provider":"aws","name":"Amazon Relational Database Service (RDS)","category":"sql","tagline":"Ready-to-use databases where AWS handles the upkeep and backups.","description":"Amazon RDS runs a database for you, so you do not have to look after the server underneath. It supports well-known database engines: MySQL, PostgreSQL, MariaDB, Oracle, SQL Server and Db2. If your app stores users, orders or payments in tables, RDS is a common home for that data, with patching, backups and failover (an automatic switch to a spare if something breaks) taken care of. Pick it when you want a standard database you already know. New accounts get a free tier for 12 months.","useCases":["Run an app's users and orders database","Move existing SQL databases to AWS","Keep a database running through failures"],"pricing":"Per instance-hour plus storage and I/O. Free tier for 12 months.","status":"ga","url":"https://aws.amazon.com/rds/","source":"https://aws.amazon.com/products/"},{"id":"aws-aurora","provider":"aws","name":"Amazon Aurora","category":"sql","tagline":"A high-performance AWS database that works with MySQL and PostgreSQL.","description":"Amazon Aurora is a database built by AWS that works with MySQL and PostgreSQL, two very popular databases, so apps written for them usually work with it. Its storage grows as you need, up to 128 TiB, and you can add up to 15 read replicas, extra copies that share the reading load. If your software product is getting busy, Aurora is built for that kind of traffic. Its Serverless v2 option adjusts capacity automatically as demand changes. Pick it over plain RDS when you expect to grow.","useCases":["Power a busy software product backend","Scale reads with up to 15 copies","Run SQL that scales capacity automatically"],"pricing":"Per instance-hour or ACU-hour plus storage and I/O.","status":"ga","url":"https://aws.amazon.com/aurora/","source":"https://aws.amazon.com/products/"},{"id":"aws-aurora-dsql","provider":"aws","name":"Amazon Aurora DSQL","category":"sql","tagline":"A database that runs in several regions at once with no servers to manage.","description":"Amazon Aurora DSQL is a database that works like PostgreSQL (a popular database) and spreads across multiple AWS Regions, meaning separate geographic locations, keeping them all in step. There are no servers for you to manage. If your app serves customers on different continents, or must stay online even if one region goes down, you can read and write in more than one place and still get consistent data. Pick it over Aurora when being always on across regions matters most. You pay per processing unit plus storage.","useCases":["Serve customers across continents with one database","Keep an app online if a region fails","Read and write in several regions"],"pricing":"Per distributed processing unit plus storage.","status":"ga","url":"https://aws.amazon.com/rds/aurora/dsql/","source":"https://aws.amazon.com/products/"},{"id":"aws-dynamodb","provider":"aws","name":"Amazon DynamoDB","category":"nosql","also":["kv-cache"],"tagline":"A fast database with no servers, great for simple lookups.","description":"Amazon DynamoDB is a database that stores simple records you look up by a key, with no servers for you to run. If your app needs to remember logged-in sessions, track game leaderboards or collect readings from connected devices, it fits well. It answers in a few thousandths of a second at almost any size, and global tables can copy data across regions. It is the default NoSQL database (one that does not use traditional tables) on AWS. You pay per request or reserve capacity, and there is a free tier.","useCases":["Store login sessions for web apps","Track game leaderboards and scores","Collect readings from connected devices"],"pricing":"On-demand per request or provisioned capacity. Free tier included.","status":"ga","url":"https://aws.amazon.com/dynamodb/","source":"https://aws.amazon.com/products/"},{"id":"aws-documentdb","provider":"aws","name":"Amazon DocumentDB (with MongoDB compatibility)","category":"nosql","tagline":"A managed database for JSON-style documents that works with MongoDB code.","description":"Amazon DocumentDB (with MongoDB compatibility) stores information as flexible documents, like a profile with whatever fields each user needs, and it speaks the same language as MongoDB, a popular document database. If your app keeps user profiles, product listings or settings as JSON (a common text format for structured data), this gives you a place to put them without running the database yourself. It keeps copies of your data and backups on storage AWS built for it. Pick it when you already have an app made for MongoDB and want to move it to AWS with little rewriting.","useCases":["Store JSON-style documents in your app","Move a MongoDB database to AWS","Run MongoDB apps with less upkeep"],"pricing":"Per instance-hour plus storage and I/O.","status":"ga","url":"https://aws.amazon.com/documentdb/","source":"https://aws.amazon.com/products/"},{"id":"aws-keyspaces","provider":"aws","name":"Amazon Keyspaces (for Apache Cassandra)","category":"nosql","tagline":"A Cassandra-style database you don't have to set up or babysit.","description":"Amazon Keyspaces (for Apache Cassandra) is a database that works with Apache Cassandra, an open source database built for huge amounts of data, without you running any servers. If you collect a steady stream of records, such as sensor readings or activity events, you can store them here and read them back by time. You can pay per read and write as they happen, or reserve capacity ahead of time. Choose it when you already have Cassandra code or skills and want AWS to handle the cluster work.","useCases":["Move Cassandra databases to AWS","Store time series records at scale","Run Cassandra-style databases without servers"],"pricing":"Per read and write request or provisioned capacity plus storage.","status":"ga","url":"https://aws.amazon.com/keyspaces/","source":"https://aws.amazon.com/products/"},{"id":"aws-neptune","provider":"aws","name":"Amazon Neptune","category":"nosql","tagline":"A database for data full of connections, like who knows whom or buys what.","description":"Amazon Neptune is a database built for relationships between things, such as people, accounts and products, stored as a graph (dots joined by lines). If you are building a recommendation feature or checking for groups of linked accounts that look like fraud, it follows those links quickly where a regular table-style database gets awkward. It supports the common graph query languages Gremlin, openCypher and SPARQL. Neptune Analytics adds graph number-crunching and vector search, which finds items with similar meaning and is handy for AI features. Pick it when the connections matter as much as the data.","useCases":["Map connections like fraud rings","Build knowledge graphs of linked facts","Power recommendations from relationships"],"pricing":"Per instance-hour plus storage and I/O.","status":"ga","url":"https://aws.amazon.com/neptune/","source":"https://aws.amazon.com/products/"},{"id":"aws-timestream","provider":"aws","name":"Amazon Timestream for InfluxDB","category":"nosql","also":["observability"],"tagline":"A managed database for measurements that change over time, like sensor data.","description":"Amazon Timestream for InfluxDB runs InfluxDB, a popular open source database for time series data (readings stamped with the moment they were taken), so you don't have to manage it yourself. If you have devices sending temperature readings, or an app that reports its health every few seconds, this stores them and lets you chart and watch them live. Pick it when you want InfluxDB specifically and would like AWS to handle the setup. The older LiveAnalytics engine is closed to new customers.","useCases":["Store IoT device metrics over time","Monitor live operational telemetry","Monitor systems in real time"],"pricing":"Per instance-hour plus storage.","status":"ga","url":"https://aws.amazon.com/timestream/","source":"https://aws.amazon.com/products/"},{"id":"aws-elasticache","provider":"aws","name":"Amazon ElastiCache","category":"kv-cache","tagline":"Keep hot data in memory so your app loads faster.","description":"Amazon ElastiCache is a fast memory-based store that sits in front of your database so your app can answer common requests almost instantly. If many people see the same popular products or leaderboard, or your app needs to remember who is logged in (sessions), you can keep that data here instead of asking the database every time. It runs Valkey and Redis OSS (two popular open source in-memory stores) or Memcached (a simpler cache), and it has a serverless option. Pick it when speed matters more than permanent storage.","useCases":["Speed up a slow database with caching","Hold user sessions in memory","Power live leaderboards"],"pricing":"Per node-hour or serverless per GB-hour and ECPU.","status":"ga","url":"https://aws.amazon.com/elasticache/","source":"https://aws.amazon.com/products/"},{"id":"aws-memorydb","provider":"aws","name":"Amazon MemoryDB","category":"kv-cache","tagline":"A very fast in-memory database that keeps your data safe, not just cached.","description":"Amazon MemoryDB keeps everything in memory for very fast responses, while also writing every change to a log spread across several data centers so nothing is lost. If you're building a live leaderboard or an app that must respond in a blink, you can use it as your main database instead of putting a cache in front of another one. It works with Valkey and Redis OSS, two popular open source in-memory stores. Choose it over ElastiCache when the data must be durable, meaning it survives failures.","useCases":["Use as a fast primary database","Serve low-latency apps from memory","Store data in memory without losing it"],"pricing":"Per node-hour plus data written.","status":"ga","url":"https://aws.amazon.com/memorydb/","source":"https://aws.amazon.com/products/"},{"id":"aws-oracle-database-at-aws","provider":"aws","name":"Oracle Database@AWS","category":"sql","tagline":"Oracle's Exadata database hardware, run inside AWS data centers.","description":"Oracle Database@AWS puts Oracle's Exadata, specialized high-end database hardware, inside AWS data centers, run by Oracle and linked directly to your private network on AWS (your VPC). If your company already depends on Oracle databases and is building new apps on AWS, this keeps the two close together so they talk quickly. You get Exadata without owning a data center. Pick it when you need Oracle specifically. Billing is usage-based, through AWS Marketplace and Oracle.","useCases":["Move Oracle databases closer to apps","Run Exadata without your own data center","Connect Oracle systems to AWS networks"],"pricing":"Usage-based via AWS Marketplace and Oracle.","status":"ga","url":"https://aws.amazon.com/oracle/","source":"https://aws.amazon.com/products/"},{"id":"aws-redshift","provider":"aws","name":"Amazon Redshift","category":"warehouse","tagline":"A data warehouse for running reports across huge amounts of business data.","description":"Amazon Redshift is a data warehouse, a database built for analyzing large amounts of data rather than running your live app. If you want dashboards that answer questions like which campaigns brought the most sales this quarter, you load your data here and query it with SQL, the standard language for asking databases questions. It can also query files in S3 (AWS file storage) and pull in data from the AWS databases Aurora and DynamoDB automatically. Pick it when you have lots of data and regular reporting. You can run it serverless or on servers you size.","useCases":["Run company-wide reports and dashboards","Combine many data sources into one warehouse","Query S3 data without moving it"],"pricing":"Per RPU-hour (serverless) or node-hour plus storage.","status":"ga","url":"https://aws.amazon.com/redshift/","source":"https://aws.amazon.com/products/"},{"id":"aws-athena","provider":"aws","name":"Amazon Athena","category":"warehouse","tagline":"Ask questions of files in S3 using SQL, with no servers to run.","description":"Amazon Athena lets you run SQL queries (the standard language for asking databases questions) directly on files sitting in S3, AWS's file storage. If you have a pile of log files or exported data and just want to answer a quick question, you point Athena at the folder and ask, with no database to set up. It runs Trino and Spark, two popular open source query engines. Pick it over a data warehouse when you query only now and then. For SQL you pay per terabyte of data scanned.","useCases":["Answer one-off questions about stored data","Analyze log files without servers","Run big-data jobs without managing servers"],"pricing":"Per TB scanned for SQL. Per DPU-hour for Spark.","status":"ga","url":"https://aws.amazon.com/athena/","source":"https://aws.amazon.com/products/"},{"id":"aws-emr","provider":"aws","name":"Amazon EMR","category":"warehouse","also":["data-pipelines"],"tagline":"Run big data jobs on Spark and Hadoop without setting up the machines.","description":"Amazon EMR processes very large amounts of data using open source tools such as Spark, Hadoop, Trino and Flink, which split work across many machines. If you need to clean and reshape millions of records before analysis, or prepare data for training a machine learning model, EMR starts the machines, runs the job and scales up or down as needed. It can run on regular AWS servers (EC2), on EKS (a container service), or serverless. Pick it when your jobs are too big for one machine or a simple SQL query.","useCases":["Run large data processing jobs","Clean and move data at scale","Prepare data for machine learning"],"pricing":"Per instance-hour surcharge or vCPU and GB-hour on serverless.","status":"ga","url":"https://aws.amazon.com/emr/","source":"https://aws.amazon.com/products/"},{"id":"aws-opensearch-service","provider":"aws","name":"Amazon OpenSearch Service","category":"warehouse","also":["observability","vector"],"tagline":"Add search, log analysis and similarity search to your app.","description":"Amazon OpenSearch Service runs OpenSearch, an open source engine for searching and analyzing text and data. If your app needs a search box that finds the right products or articles, or you want to dig through server logs, this is the tool for it. It also supports vector search, which finds items with similar meaning and is handy for AI features. You can run your own clusters (groups of servers) or use the serverless option. Pick it over plain database search when you need fast, flexible text search.","useCases":["Add search to your website","Analyze application and server logs","Find items similar in meaning"],"pricing":"Per instance-hour plus storage, or per OCU-hour on serverless.","status":"ga","url":"https://aws.amazon.com/opensearch-service/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloudsearch","provider":"aws","name":"Amazon CloudSearch","category":"warehouse","tagline":"An older managed search service for websites and apps.","description":"Amazon CloudSearch is a managed service that adds a search box to a website or app, so visitors can find content by typing keywords. If your site already uses it to search your articles or products, it keeps working as before. You pay per instance-hour, meaning each hour a server runs. It is now in maintenance, and AWS recommends Amazon OpenSearch Service for new projects.","useCases":["Keep existing site search running","Maintain older search setups","Support apps already built on it"],"pricing":"Per instance-hour.","status":"closed","url":"https://aws.amazon.com/cloudsearch/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-quick","provider":"aws","name":"Amazon Quick","category":"ai-apps","also":["warehouse"],"tagline":"An AI assistant for work that also builds dashboards and simple apps.","description":"Amazon Quick is a workplace AI assistant that also turns your data into charts and dashboards, and lets you build simple apps without writing code. The older QuickSight analytics service is now part of it. If you run a small team and want to ask a question like which product sold best last month in plain English and get a chart back, this is a good fit. You can also use it to automate routine tasks. Pricing is per person per month depending on their role, with extra charges for added capacity.","useCases":["Build dashboards from company data","Ask data questions in plain English","Automate routine office tasks"],"pricing":"Per user per month by role, plus capacity add-ons.","status":"ga","url":"https://aws.amazon.com/quick/","source":"https://aws.amazon.com/products/"},{"id":"aws-datazone","provider":"aws","name":"Amazon DataZone","category":"data-pipelines","tagline":"A company catalog where teams publish data and others request access.","description":"Amazon DataZone is a shared catalog where teams publish their data and other teams can find it and ask for access. If your sales team has data that marketing wants to use, DataZone gives marketing a way to request it and gives the owners a way to approve it. It also includes a business glossary, which is a shared list of what each term means, so everyone reads the data the same way. Pick it when data sharing is getting messy across teams, rather than for one small app.","useCases":["Publish data for other teams","Request access to team data","Share a glossary of data terms"],"pricing":"Per user and metadata storage and requests.","status":"ga","url":"https://aws.amazon.com/datazone/","source":"https://aws.amazon.com/products/"},{"id":"aws-sagemaker-unified-studio","provider":"aws","name":"Amazon SageMaker","category":"ai-platform","also":["warehouse","data-pipelines"],"tagline":"One workspace for your data, analytics and AI projects.","description":"Amazon SageMaker is the newer version of AWS's machine learning product, now a single workspace for data, analytics and AI. It brings together a lakehouse (one place to keep both raw files and organized tables), analytics tools and model building. If your team wants to prepare customer data and then train a prediction model on it without hopping between tools, you would start here. It sits on top of S3 (file storage), Redshift (a data warehouse for reports) and Glue (data preparation), and you pay usage-based prices across those services.","useCases":["Prepare customer data for analysis","Train a prediction model","Manage data and AI projects together"],"pricing":"Usage-based across the underlying services.","status":"ga","url":"https://aws.amazon.com/sagemaker/","source":"https://aws.amazon.com/products/"},{"id":"aws-glue","provider":"aws","name":"AWS Glue","category":"data-pipelines","tagline":"Move and tidy up data between your tools, with no servers to manage.","description":"AWS Glue finds your data, cleans it up and moves it where it needs to go, without you managing any servers. If your app stores orders in one place and you want them organized for reporting in another, you can build a job in Spark or Python (a big-data tool and a popular programming language) to do it. Glue also keeps a catalog, an index of what data you have and how it is laid out, that Athena, Redshift and EMR (other AWS tools for searching and processing data) all read from. That makes it the natural pick if you use those. Jobs are billed by DPU-hour, a unit of computing time.","useCases":["Move and organize data between systems","Build an index of stored data","Detect how data is structured"],"pricing":"Per DPU-hour for jobs plus catalog storage and requests.","status":"ga","url":"https://aws.amazon.com/glue/","source":"https://aws.amazon.com/products/"},{"id":"aws-glue-databrew","provider":"aws","name":"AWS Glue DataBrew","category":"data-pipelines","tagline":"Clean up messy data by pointing and clicking, with no code.","description":"AWS Glue DataBrew is a point-and-click tool for cleaning up messy data. If you have a spreadsheet export with inconsistent dates, duplicate names and blank cells, you can fix it by choosing from more than 250 built-in steps instead of writing code. It is handy for preparing data before you feed it to a machine learning model. Pick it over Glue's code-based jobs when the person doing the cleanup is an analyst rather than an engineer. You pay per node-hour, which is the time the tool spends running your jobs or sessions.","useCases":["Clean messy spreadsheets without coding","Fix duplicates, blanks and bad dates","Prepare data for machine learning"],"pricing":"Per node-hour for jobs and interactive sessions.","status":"ga","url":"https://aws.amazon.com/glue/features/databrew/","source":"https://aws.amazon.com/products/"},{"id":"aws-lake-formation","provider":"aws","name":"AWS Lake Formation","category":"data-pipelines","also":["management"],"tagline":"Set up a data lake and control who can see which data.","description":"AWS Lake Formation helps you set up a data lake, which is a big storage area for all your company's data, on S3 (AWS file storage). Its main strength is control over who sees what, down to specific columns and rows. If you keep customer data and want your marketing team to see purchases but not email addresses, you can set that rule once. The rules apply across Athena, Redshift Spectrum and EMR, which are AWS tools for searching and processing data. Lake Formation itself costs nothing, and you pay only for the other services it works with.","useCases":["Set up a central data storage area","Limit who sees certain columns","Control access to specific rows"],"pricing":"No charge for Lake Formation. Pay for integrated services.","status":"ga","url":"https://aws.amazon.com/lake-formation/","source":"https://aws.amazon.com/products/"},{"id":"aws-data-pipeline","provider":"aws","name":"AWS Data Pipeline","category":"data-pipelines","tagline":"An older scheduler for moving data on a timetable.","description":"AWS Data Pipeline is a scheduling service that moves and transforms data on a timetable, such as copying yesterday's records into a report every morning. It suits older-style data jobs, and you pay per pipeline activity per month. For anything new, you will likely want a different tool. This service is now legacy: AWS no longer adds features and points people to Step Functions (a tool for chaining tasks together) and MWAA (managed Airflow, a popular workflow scheduler) instead.","useCases":["Run older scheduled data copies","Copy records into reports daily","Maintain existing legacy data jobs"],"pricing":"Per pipeline activity per month.","status":"deprecated","url":"https://aws.amazon.com/datapipeline/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-appflow","provider":"aws","name":"Amazon AppFlow","category":"data-pipelines","tagline":"Copy data between apps like Salesforce and AWS without code.","description":"Amazon AppFlow moves data between the business apps you already use and AWS, without writing code. If your sales team lives in Salesforce and you want those records copied into S3 (AWS file storage) for reporting, you set up a flow, choose a schedule and it runs. It connects to apps such as Slack, SAP and ServiceNow too, and can reshape the data along the way. Pick it over custom code when you just need a reliable sync between two tools. You pay per flow run plus the amount of data processed.","useCases":["Copy Salesforce records into storage","Sync data between business apps","Reshape data while it moves"],"pricing":"Per flow run plus data processed.","status":"ga","url":"https://aws.amazon.com/appflow/","source":"https://aws.amazon.com/products/"},{"id":"aws-data-exchange","provider":"aws","name":"AWS Data Exchange","category":"data-pipelines","tagline":"Buy or sell datasets from outside companies in one marketplace.","description":"AWS Data Exchange is a marketplace where you can subscribe to data sold by other companies, such as market data. If you are building a product that needs outside data and you do not want to collect it yourself, you can subscribe and have it delivered to S3 (AWS file storage), Redshift (a data warehouse) or through APIs, which let programs request data automatically. You can also publish your own data products for others to buy. Providers set the subscription prices, so costs vary by dataset.","useCases":["Buy market data from outside providers","Sell your own datasets to others","Deliver purchased data to storage"],"pricing":"Subscription prices set by providers.","status":"ga","url":"https://aws.amazon.com/data-exchange/","source":"https://aws.amazon.com/products/"},{"id":"aws-clean-rooms","provider":"aws","name":"AWS Clean Rooms","category":"data-pipelines","also":["warehouse"],"tagline":"Analyze data with a partner without either side seeing the raw records.","description":"AWS Clean Rooms lets two or more companies analyze their data together without handing over the underlying records. If you are a brand and a retail partner who want to see which ads led to purchases, each of you keeps your own data and controls what the other can see. It also supports joint machine learning, where a prediction model is trained on the combined data. Pick it over swapping spreadsheets when privacy matters. You pay for the queries and machine learning jobs you run.","useCases":["Measure ad results with a partner","Analyze partner data without sharing records","Train models on shared data"],"pricing":"Per compute query and ML job usage.","status":"ga","url":"https://aws.amazon.com/clean-rooms/","source":"https://aws.amazon.com/products/"},{"id":"aws-entity-resolution","provider":"aws","name":"AWS Entity Resolution","category":"data-pipelines","also":["ai-apps"],"tagline":"Spot duplicate or related records, like the same customer, across systems.","description":"AWS Entity Resolution finds records that refer to the same person or business, even when they are spelled or stored differently. If a customer shows up in your store database, your email tool and your support system, it can link those entries so you see one customer instead of three. You can match using your own rules, machine learning or a data provider's matching service. It works without moving your data out of where it already lives. Pricing is per 1,000 records processed.","useCases":["Match the same customer across systems","Remove duplicate customer records","Build one view per customer"],"pricing":"Per 1,000 records processed.","status":"ga","url":"https://aws.amazon.com/entity-resolution/","source":"https://aws.amazon.com/products/"},{"id":"aws-finspace","provider":"aws","name":"Amazon FinSpace","category":"specialized","also":["warehouse"],"tagline":"Store and analyze financial data, including fast market time series.","description":"Amazon FinSpace is a data service built for financial companies. It stores and prepares financial data, and offers managed kdb Insights, a database tool for time series, meaning data that changes moment to moment, like stock prices. If you are building a trading research tool and need to dig through years of market data, this is aimed at you. It is a specialist pick rather than a general-purpose database. Pricing is usage-based, and you contact sales for kdb.","useCases":["Analyze years of market data","Research trading strategies with data","Store and prepare financial data"],"pricing":"Usage-based. Contact sales for kdb.","status":"ga","url":"https://aws.amazon.com/finspace/","source":"https://aws.amazon.com/products/"},{"id":"aws-kinesis","provider":"aws","name":"Amazon Kinesis Data Streams","category":"queues","also":["data-pipelines"],"tagline":"Catch a nonstop flow of events and let many apps read them in real time.","description":"Amazon Kinesis Data Streams collects a constant flow of data, like clicks, logs or sensor readings, and holds onto it so other programs can read it right away. If your app tracks every click on your website, you can send each click here and have one program build live dashboards while another saves everything for later. Many readers can use the same stream at once, and the on-demand option grows on its own. Pick it when you need to handle a heavy flow of events as they happen.","useCases":["Capture clickstream events from users","Stream logs and events in real time","Collect sensor data from devices"],"pricing":"Per shard-hour or per GB on-demand.","status":"ga","url":"https://aws.amazon.com/kinesis/","source":"https://aws.amazon.com/products/"},{"id":"aws-kinesis-video-streams","provider":"aws","name":"Amazon Kinesis Video Streams","category":"video","also":["iot"],"tagline":"Send video from cameras and devices to the cloud, then store or replay it.","description":"Amazon Kinesis Video Streams takes video from cameras and other devices, stores it, and lets you play it back later. If you are building a home security app or a robot that needs to see, each device can send its video here instead of you building your own video pipeline. It also supports WebRTC, a technology for live two-way video, so people can talk face to face through a device. Choose it over a general data stream when your data is video.","useCases":["Stream video from security cameras","Stream video for smart home devices","Send video from robots for analysis"],"pricing":"Per GB ingested, stored and consumed.","status":"ga","url":"https://aws.amazon.com/kinesis/video-streams/","source":"https://aws.amazon.com/products/"},{"id":"aws-data-firehose","provider":"aws","name":"Amazon Data Firehose","category":"data-pipelines","also":["queues"],"tagline":"Automatically deliver streaming data into storage and analytics tools.","description":"Amazon Data Firehose picks up streaming data, groups it into batches, optionally reshapes it, and drops it into places like S3 (file storage), Redshift (a data warehouse for analysis) or OpenSearch (a search and analytics tool). If your app produces logs you want to keep and search later, you point them at Firehose and it delivers them for you. You do not need to write code to run it. Pick it when you mainly need data moved into storage, not processed in complicated ways.","useCases":["Stream logs into storage automatically","Load live data into warehouses","Convert streaming data formats on the way"],"pricing":"Per GB ingested plus format conversion.","status":"ga","url":"https://aws.amazon.com/firehose/","source":"https://aws.amazon.com/products/"},{"id":"aws-managed-flink","provider":"aws","name":"Amazon Managed Service for Apache Flink","category":"data-pipelines","also":["queues"],"tagline":"Analyze live data as it arrives, without running the servers yourself.","description":"Amazon Managed Service for Apache Flink runs Flink, a popular open source tool for crunching data as it flows in, without you managing the machines underneath. If you want to spot unusual payments the moment they happen, or keep a running total of sales per minute, you write a Flink application and AWS runs it. It suits jobs like live totals, real-time ETL (cleaning and moving data) and anomaly detection. Pick it when simply moving data is not enough and you need to calculate on it in motion.","useCases":["Analyze data streams as they arrive","Spot unusual activity in real time","Compute running totals from live events"],"pricing":"Per KPU-hour.","status":"ga","url":"https://aws.amazon.com/managed-service-apache-flink/","source":"https://aws.amazon.com/products/"},{"id":"aws-msk","provider":"aws","name":"Amazon Managed Streaming for Apache Kafka (MSK)","category":"queues","tagline":"Run Apache Kafka, a popular event streaming tool, without managing it.","description":"Amazon MSK is Apache Kafka, a widely used system for passing streams of events between apps, run for you by AWS. If your product has many services that need to share events, like orders placed or accounts updated, Kafka acts as the central pipe they all read from and write to. MSK recovers automatically from problems and has a serverless option, so you can skip sizing machines yourself. Pick it over Kinesis when your team already knows Kafka or wants to keep tools built for it.","useCases":["Let separate services exchange messages","Stream events between app services","Track database changes as they happen"],"pricing":"Per broker-hour plus storage, or serverless per cluster-hour and partition.","status":"ga","url":"https://aws.amazon.com/msk/","source":"https://aws.amazon.com/products/"},{"id":"aws-sqs","provider":"aws","name":"Amazon Simple Queue Service (SQS)","category":"queues","tagline":"A waiting line for tasks, so your app can handle work at its own pace.","description":"Amazon SQS is a message queue, a waiting line where one part of your app leaves tasks for another part to pick up. If users upload videos and each one needs processing, you drop a note in the queue and a worker handles it when ready, so a sudden rush does not overwhelm anything. It offers standard queues and FIFO queues, which keep messages in order, and it grows without capacity planning. It is a simple default for passing work along, with 1 million free requests each month.","useCases":["Queue background jobs for processing","Decouple services so they run separately","Absorb sudden spikes in traffic"],"pricing":"Per million requests. Free tier: 1M requests monthly.","status":"ga","url":"https://aws.amazon.com/sqs/","source":"https://aws.amazon.com/products/"},{"id":"aws-sns","provider":"aws","name":"Amazon Simple Notification Service (SNS)","category":"queues","also":["messaging"],"tagline":"Send one message to many places: apps, emails, texts and phone alerts.","description":"Amazon SNS sends a message once and delivers it to many receivers, including queues, Lambda functions (small bits of code that run on demand), web addresses, text messages and email. If a customer places an order, you can publish one event and have it email the customer, alert your team and start processing at the same time. It also sends mobile push notifications. Pick it when one event needs to reach many places, whereas a queue like SQS hands each task to one worker.","useCases":["Send one message to many subscribers","Send alerts by text or email","Push notifications to mobile apps"],"pricing":"Per million publishes plus delivery charges.","status":"ga","url":"https://aws.amazon.com/sns/","source":"https://aws.amazon.com/products/"},{"id":"aws-eventbridge","provider":"aws","name":"Amazon EventBridge","category":"queues","also":["workflows"],"tagline":"Route events between your apps and services, or run tasks on a schedule.","description":"Amazon EventBridge is a central switchboard that takes in events, meaning notes that something happened, and sends them to the right place based on rules you set. If a partner tool you use reports that a customer just paid, a rule can forward that event to your own code. It also includes a cron-style scheduler, which runs tasks at set times, such as a nightly report. Pick it when you want different pieces of your product to react to events without wiring them directly together.","useCases":["Trigger workflows when events happen","Connect SaaS tools to your app","Run tasks on a set schedule"],"pricing":"Per million events published.","status":"ga","url":"https://aws.amazon.com/eventbridge/","source":"https://aws.amazon.com/products/"},{"id":"aws-mq","provider":"aws","name":"Amazon MQ","category":"queues","tagline":"Managed ActiveMQ and RabbitMQ message brokers for apps that already use them.","description":"Amazon MQ runs ActiveMQ and RabbitMQ, two established message brokers (software that passes messages between apps), so AWS handles the running of them. If you have an existing app that already talks through JMS, AMQP or MQTT, which are standard messaging protocols, you can move it over without rewriting how it sends messages. Its main job is helping with migrations. For a brand new project, a simpler choice like SQS or SNS is often a better starting point.","useCases":["Keep existing messaging apps working","Migrate older message brokers over","Run ActiveMQ or RabbitMQ brokers"],"pricing":"Per broker-hour plus storage.","status":"ga","url":"https://aws.amazon.com/amazon-mq/","source":"https://aws.amazon.com/products/"},{"id":"aws-mwaa","provider":"aws","name":"Amazon Managed Workflows for Apache Airflow (MWAA)","category":"workflows","also":["data-pipelines"],"tagline":"Schedule and run multi-step data jobs with Airflow, managed for you.","description":"Amazon MWAA is Apache Airflow, a popular tool for scheduling and ordering data jobs written in Python, run for you by AWS. If every night you need to pull data from your app, clean it up and refresh a report, you describe those steps in Python and Airflow runs them in the right order. It also fits machine learning workflows. Pick it when your team already uses Airflow or likes defining pipelines as code.","useCases":["Schedule data pipeline jobs in Python","Keep data pipeline steps running in order","Run machine learning workflows on a schedule"],"pricing":"Per environment-hour plus workers and storage.","status":"ga","url":"https://aws.amazon.com/mwaa/","source":"https://aws.amazon.com/products/"},{"id":"aws-step-functions","provider":"aws","name":"AWS Step Functions","category":"workflows","also":["serverless"],"tagline":"Build multi-step processes as a visual flowchart, with retries built in.","description":"AWS Step Functions lets you chain steps into a workflow, see it as a flowchart, and have AWS run it for you. If your store has to charge a card, reserve stock and send a receipt, Step Functions runs each step in order, retries any that fail and can run steps in parallel. It can even pause for a human approval. It connects to over 220 AWS services, including Lambda (small bits of code that run on demand). Pick it when a process has several steps you need to track.","useCases":["Coordinate Lambda and other AWS services","Run steps with retries and parallel tasks","Pause workflows for human approval"],"pricing":"Per state transition, or per request for express workflows.","status":"ga","url":"https://aws.amazon.com/step-functions/","source":"https://aws.amazon.com/products/"},{"id":"aws-swf","provider":"aws","name":"Amazon Simple Workflow Service (SWF)","category":"workflows","tagline":"An older workflow service; Step Functions is the pick for new projects.","description":"Amazon SWF is a workflow service, a tool for coordinating work that is split across many steps or machines. It is considered a legacy service, so you will mostly meet it in older projects. If you inherit an app that already uses it, it keeps coordinating those steps as before. For anything new, Step Functions is the recommended replacement, and it gives you a visual workflow with retries and approvals built in.","useCases":["Keep existing workflow systems running","Coordinate distributed work across services","Use Step Functions for new projects"],"pricing":"Per workflow execution and task.","status":"ga","url":"https://aws.amazon.com/swf/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-b2b-data-interchange","provider":"aws","name":"AWS B2B Data Interchange","category":"data-pipelines","also":["end-user"],"tagline":"Translate partners' business documents into data your app can read.","description":"AWS B2B Data Interchange converts EDI documents, an older standard format that big companies use to swap orders and invoices, into JSON or XML, which are common formats apps can read, and back again. If you sell to a retailer or buy from suppliers who send X12 files (one popular EDI style), you can turn each incoming order into data your app understands. Pick it when you would rather not write and maintain your own converter for these formats. You pay per document transformed.","useCases":["Convert incoming EDI orders into JSON","Turn supplier X12 files into app data","Send converted data back as EDI"],"pricing":"Per document transformed.","status":"ga","url":"https://aws.amazon.com/b2b-data-interchange/","source":"https://aws.amazon.com/products/"},{"id":"aws-bedrock","provider":"aws","name":"Amazon Bedrock","category":"ai-models","also":["ai-platform","ai-gateway"],"tagline":"One door to many AI models, plus tools to build chatbots and assistants.","description":"Amazon Bedrock gives you one API, a single way for your code to make requests, to AI models from Anthropic, Amazon, Meta, Mistral and others. If you are building a support chatbot that answers from your own help articles, you can use its knowledge bases, guardrails (rules for what the AI may say) and fine-tuning (extra training on your own examples). Pick it when you want to try or switch between models without rewiring your app. Pricing is per token (a small chunk of text) or reserved capacity, and varies by model.","useCases":["Build chatbots that answer from your articles","Switch AI models without rewriting your app","Summarize documents with AI models"],"pricing":"Per token or provisioned throughput, varies by model.","status":"ga","url":"https://aws.amazon.com/bedrock/","source":"https://aws.amazon.com/products/"},{"id":"aws-bedrock-agentcore","provider":"aws","name":"Amazon Bedrock AgentCore","category":"agents","also":["ai-platform"],"tagline":"Run AI agents in production, with memory, tools and security built in.","description":"Amazon Bedrock AgentCore is a set of ready-made parts for running AI agents, which are programs that use an AI model to carry out tasks. If you have built an agent that books meetings or researches leads, it gives you a place to run it, remember past conversations, control who it acts as, and connect it to outside tools. It also includes a web browser and a code runner for agents to use. Pick it when you want to skip building that plumbing yourself. It works with any framework or model, and you pay for what each part uses.","useCases":["Host AI agents in production","Connect agents to outside tools","Let agents browse the web"],"pricing":"Usage-based per component, such as runtime CPU and memory seconds.","status":"ga","url":"https://aws.amazon.com/bedrock/agentcore/","source":"https://aws.amazon.com/products/"},{"id":"aws-nova","provider":"aws","name":"Amazon Nova","category":"ai-models","tagline":"Amazon's own AI models for text, images, video and speech, at lower cost.","description":"Amazon Nova is Amazon's own family of AI models, covering text, images, video and speech, and you use it through Amazon Bedrock. If your app writes product descriptions, summarizes documents or needs to understand pictures and video, Nova can handle those jobs. It costs less than many third-party models, so it suits high-volume work where the bill matters. You can also customize it through Bedrock, and you pay per token, a small chunk of text, as with other Bedrock models.","useCases":["Generate text for high-volume jobs","Write product descriptions at lower cost","Understand images and video in your app"],"pricing":"Per token via Bedrock.","status":"ga","url":"https://aws.amazon.com/nova/","source":"https://aws.amazon.com/products/"},{"id":"aws-nova-act","provider":"aws","name":"Amazon Nova Act","category":"agents","also":["ai-apps"],"tagline":"AI agents that click, type and navigate websites to do tasks for you.","description":"Amazon Nova Act is a service and toolkit for building AI agents that use a web browser the way a person would, by clicking, typing and moving between pages. If you need to test your app's signup flow every day, fill in the same forms again and again, or collect information from sites with no easy data feed, you can hand that routine to an agent. Pick it when the task lives behind a website's screens and there is no simpler way in. You can run many agents at once, and pricing is per agent hour.","useCases":["Test signup flows in a browser daily","Fill in the same web forms","Collect data from sites with no feed"],"pricing":"Usage-based per agent hour.","status":"ga","url":"https://aws.amazon.com/nova/act/","source":"https://aws.amazon.com/products/"},{"id":"aws-claude-platform-on-aws","provider":"aws","name":"Claude Platform on AWS","category":"ai-models","tagline":"Use Anthropic's Claude directly, billed and managed through your AWS account.","description":"Claude Platform on AWS lets you use Anthropic's own Claude API, the same one Anthropic offers directly, while billing and access control run through your AWS account. If your company already buys software through AWS and wants to add a Claude-powered assistant to your product, this keeps everything on one invoice and one set of permissions. Pick it when you want Anthropic's native platform features but need AWS contracts and security rules to stay in charge. Pricing follows Anthropic's token rates, billed through AWS.","useCases":["Use Claude under your AWS contracts","Add a Claude assistant on one invoice","Keep AWS security rules for Claude access"],"pricing":"Anthropic token pricing billed through AWS.","status":"ga","url":"https://aws.amazon.com/claude-platform/","source":"https://aws.amazon.com/products/"},{"id":"aws-sagemaker-ai","provider":"aws","name":"Amazon SageMaker AI","category":"ai-platform","tagline":"Build, train and host your own machine learning models.","description":"Amazon SageMaker AI is a workspace for building machine learning models, programs that learn patterns from data. If you want to train a model on your own customer data to predict who might cancel, it gives you notebooks to experiment in, training on many machines at once, and endpoints (always-on addresses) to serve the finished model. Pick it when you need your own model rather than a ready-made one. Some parts, such as Ground Truth and Clarify, are in maintenance. You pay per instance-hour, meaning each hour a server runs.","useCases":["Train a custom model on your data","Host finished models on always-on endpoints","Experiment in notebooks before training"],"pricing":"Per instance-hour for notebooks, training and endpoints.","status":"ga","url":"https://aws.amazon.com/sagemaker-ai/","source":"https://aws.amazon.com/products/"},{"id":"aws-a2i","provider":"aws","name":"Amazon Augmented AI (A2I)","category":"ai-apps","also":["ai-platform"],"tagline":"Send uncertain AI answers to a person to double-check.","description":"Amazon Augmented AI, or A2I, adds human reviewers to your AI workflow. It routes low-confidence predictions, the ones the AI is unsure about, to a person who can check them. If your app reads uploaded documents or screens user posts, you can let the AI handle the clear cases and send only the doubtful ones to a reviewer. Pick it when mistakes are costly and you want a safety net without reviewing everything by hand. You pay per human review loop.","useCases":["Send doubtful AI results to reviewers","Review uncertain document reads by hand","Add human checks to content moderation"],"pricing":"Per human review loop.","status":"ga","url":"https://aws.amazon.com/augmented-ai/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-kendra","provider":"aws","name":"Amazon Kendra","category":"ai-search","tagline":"Search across company documents with AI (older service).","description":"Amazon Kendra is a search service that finds answers in company documents, with connectors that pull in content from the places teams already store it. If an organization already runs its internal search on Kendra, it keeps working as before. Pricing is per index-hour (an index is the searchable catalog of your content) plus connectors. AWS lists Kendra as closed and in maintenance, and suggests Bedrock Knowledge Bases for new builds.","useCases":["Keep existing internal search running","Search company documents with AI","Pull content from connected sources"],"pricing":"Per index-hour plus connectors.","status":"closed","url":"https://aws.amazon.com/kendra/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-comprehend","provider":"aws","name":"Amazon Comprehend","category":"ai-apps","tagline":"Read text and pick out names, sentiment and personal details.","description":"Amazon Comprehend reads text and tells you what is in it: people and places, key phrases, whether the tone is positive or negative, and personal information (PII, such as names and addresses). If your app collects customer reviews or support messages, you can use it to spot unhappy customers or hide private details before storing the text. Pick it when you want quick text analysis without training your own model. Topic modeling and event detection features are in maintenance. Pricing is per 100 characters analyzed.","useCases":["Spot unhappy customers in reviews","Hide personal details before storing text","Sort documents into categories"],"pricing":"Per 100 characters of text analyzed.","status":"ga","url":"https://aws.amazon.com/comprehend/","source":"https://aws.amazon.com/products/"},{"id":"aws-comprehend-medical","provider":"aws","name":"Amazon Comprehend Medical","category":"ai-apps","tagline":"Pull conditions, medications and dosages out of clinical notes.","description":"Amazon Comprehend Medical reads medical text and picks out conditions, medications and dosages. It also links them to standard code systems such as ICD-10 (for diagnoses) and RxNorm (for drugs). If you are building a health app that handles doctors' notes, you can turn messy paragraphs into organized data, which helps with record searching or medical coding. Pick it over a general text tool when your text is clinical and the medical terms matter. Pricing is per unit of text analyzed.","useCases":["Pull conditions and medications from notes","Organize clinical notes into searchable data","Link medical terms to standard codes"],"pricing":"Per unit of text analyzed.","status":"ga","url":"https://aws.amazon.com/comprehend/medical/","source":"https://aws.amazon.com/products/"},{"id":"aws-lex","provider":"aws","name":"Amazon Lex","category":"ai-apps","tagline":"Build chatbots that talk or type, using the tech behind Alexa.","description":"Amazon Lex lets you build chatbots that people can type or speak to, using the same conversation technology as Alexa. You define what people want (intents) and the details you need from them (slots), such as a date or an order number. If you run a support line and want a bot to answer common questions before a human steps in, Lex fits well. Pick it when you use Amazon Connect, AWS's call center service, since the two work together. You pay per text or speech request.","useCases":["Build chatbots people can type or speak to","Answer common support questions before handoff","Add bots to Amazon Connect call lines"],"pricing":"Per text or speech request.","status":"ga","url":"https://aws.amazon.com/lex/","source":"https://aws.amazon.com/products/"},{"id":"aws-polly","provider":"aws","name":"Amazon Polly","category":"ai-apps","tagline":"Turn written text into natural-sounding spoken audio.","description":"Amazon Polly reads your text out loud using lifelike computer voices in many languages. If you are building an app that narrates articles, greets phone callers with spoken menus, or offers audio for people who find screens hard to read, Polly makes the audio for you. It offers neural and generative voices, which are newer AI-made voices that sound more human. Pricing is per million characters of text, with a free tier for the first 12 months, so it is cheap to try.","useCases":["Turn text into spoken narration","Generate phone menu voice prompts","Create audio for accessibility"],"pricing":"Per million characters. Free tier for 12 months.","status":"ga","url":"https://aws.amazon.com/polly/","source":"https://aws.amazon.com/products/"},{"id":"aws-transcribe","provider":"aws","name":"Amazon Transcribe","category":"ai-apps","tagline":"Turn recordings and live audio into written text.","description":"Amazon Transcribe listens to audio and writes down what was said. If your product records sales calls, meetings or videos, you can use it to create transcripts, subtitles or searchable notes. It can label who is speaking, hide sensitive details such as names or card numbers, and handle live audio as it streams. Versions built for medical conversations and call center analysis also exist. You pay per second of audio, and there is a free tier for 12 months.","useCases":["Turn audio recordings into text","Create subtitles from video audio","Take notes from recorded meetings"],"pricing":"Per second of audio. Free tier for 12 months.","status":"ga","url":"https://aws.amazon.com/transcribe/","source":"https://aws.amazon.com/products/"},{"id":"aws-translate","provider":"aws","name":"Amazon Translate","category":"ai-apps","tagline":"Translate text and documents between many languages automatically.","description":"Amazon Translate converts text from one language into another using AI. If you want your app, help pages or marketing content to reach people in other countries, or you want a chat feature where two people can talk in different languages, you send the text to Translate and get it back translated. Its custom terminology support lets you decide how your brand names or product terms are translated, so they do not get mangled. Pricing is per million characters.","useCases":["Translate website content into many languages","Translate chat messages in real time","Keep custom terms consistent across translations"],"pricing":"Per million characters.","status":"ga","url":"https://aws.amazon.com/translate/","source":"https://aws.amazon.com/products/"},{"id":"aws-textract","provider":"aws","name":"Amazon Textract","category":"ai-apps","tagline":"Pull text, form fields and tables out of scanned documents.","description":"Amazon Textract reads documents and photos of documents and pulls out the information inside. Basic text recognition (OCR) only gives you loose words, but Textract also understands form fields, tables, IDs and invoices. If your app lets customers upload invoices or sign-up forms and you want the amounts and names filled in automatically, it saves you from typing them by hand or building your own reader. It is priced per page, depending on which features you use.","useCases":["Pull data from invoices automatically","Turn paper forms into digital fields","Read details from ID documents"],"pricing":"Per page by feature.","status":"ga","url":"https://aws.amazon.com/textract/","source":"https://aws.amazon.com/products/"},{"id":"aws-rekognition","provider":"aws","name":"Amazon Rekognition","category":"ai-apps","tagline":"Find objects, faces, text and unsafe content in photos and video.","description":"Amazon Rekognition looks at images and video and tells you what is in them. If your app lets people upload photos, you can use it to flag unsafe content before it goes public, check that a selfie matches a face in another photo, or spot objects and text in a picture. It is a ready-made option, so you do not need to train your own image model. Some streaming and batch moderation features are in maintenance, so check which ones you plan to rely on. Pricing is per image or per minute of video.","useCases":["Compare faces for identity checks","Screen images for unsafe content","Detect objects in photos and video"],"pricing":"Per image or per minute of video.","status":"ga","url":"https://aws.amazon.com/rekognition/","source":"https://aws.amazon.com/products/"},{"id":"aws-personalize","provider":"aws","name":"Amazon Personalize","category":"ai-apps","tagline":"Show each customer recommendations based on what they do.","description":"Amazon Personalize gives every user their own suggestions, like products they might buy or content they might enjoy next. You send it records of what people click, buy or watch, and it builds a recommendation system from that, with no machine learning expertise needed. If you run a store, a media app or an email program and want a \"you might like\" feature, it is far quicker than building one yourself. You pay per GB of data sent in, per training hour and per request.","useCases":["Recommend products to each shopper","Personalize email content for each user","Suggest the next best action"],"pricing":"Per GB ingested, training hour and request.","status":"ga","url":"https://aws.amazon.com/personalize/","source":"https://aws.amazon.com/products/"},{"id":"aws-forecast","provider":"aws","name":"Amazon Forecast","category":"ai-apps","tagline":"Predict future numbers like demand from past data (legacy).","description":"Amazon Forecast predicts future numbers, such as next month's sales, by studying how those numbers changed in the past. If you already have demand forecasts running on it, it keeps producing them, and you pay per forecast generated. For a new project, AWS points to SageMaker Canvas, a point-and-click tool for building predictions, instead. The service is now in maintenance, so it is not the place to start something new.","useCases":["Keep existing demand forecasts running","Generate forecasts from time series data","Use SageMaker Canvas for new work"],"pricing":"Per forecast generated.","status":"closed","url":"https://aws.amazon.com/forecast/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-fraud-detector","provider":"aws","name":"Amazon Fraud Detector","category":"bots","also":["ai-apps"],"tagline":"Spot likely fraud in signups and payments (closed to new customers).","description":"Amazon Fraud Detector scores events like new signups and payments to estimate how likely each one is to be fraud, using machine learning. If you run a shop or app and want to catch stolen cards or hijacked accounts, that is the kind of check it was built for, and it charges per prediction. Its uses include account takeover detection and payment fraud. The service is in maintenance and closed to new customers, so you cannot start using it now.","useCases":["Score signups for fraud","Score payments for fraud","Flag account takeover attempts"],"pricing":"Per prediction.","status":"closed","url":"https://aws.amazon.com/fraud-detector/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-q-developer","provider":"aws","name":"Amazon Q Developer","category":"devtools","also":["ai-apps"],"tagline":"An AI helper that writes code, explains errors and fixes AWS problems.","description":"Amazon Q Developer is an AI assistant that works inside your code editor, the command line and the AWS console, the website where you manage your AWS account. If you are building on AWS and get stuck on an error message, you can ask it what went wrong, or have it write code for you. It can also upgrade older Java versions in existing code. Pick it over a general chatbot when your questions are about AWS itself. There is a free tier, and Pro costs about $19 per user per month.","useCases":["Write code with an AI assistant","Explain errors and troubleshoot AWS issues","Upgrade Java versions in code"],"pricing":"Free tier. Pro at about $19 per user per month.","status":"ga","url":"https://aws.amazon.com/q/developer/","source":"https://aws.amazon.com/products/"},{"id":"aws-q-business","provider":"aws","name":"Amazon Q Business","category":"ai-search","also":["ai-apps"],"tagline":"AI assistant that answers questions from company data (legacy).","description":"Amazon Q Business is an AI assistant that answers questions using a company's own data sources. A team that wants staff to ask \"what is our refund policy?\" and get an answer drawn from company files could use it, and it is billed per user per month. Teams already using it for internal question answering can keep doing so. The service is in maintenance, and Amazon Quick is its successor, so new projects should look there.","useCases":["Keep existing company Q&A running","Answer questions from company data","Use Amazon Quick for new projects"],"pricing":"Per user per month.","status":"closed","url":"https://aws.amazon.com/q/business/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-kiro","provider":"aws","name":"Kiro","category":"devtools","also":["agents"],"tagline":"An AI coding app that turns your ideas into plans, tasks and code.","description":"Kiro is an AI coding environment, an app where you write and manage code, built around AI agents that do work for you. You describe what you want, and it turns your prompt into a written spec (a plan of what to build), then a list of tasks, then code. If you are adding a new feature and want the AI to plan before it types, that is what sets it apart. Hooks and steering files let you set automatic actions and written guidelines for the AI. There is a free tier plus paid per-user plans.","useCases":["Turn prompts into specs and tasks","Build features from written specs","Guide the AI with steering files"],"pricing":"Free tier plus paid per-user plans.","status":"ga","url":"https://kiro.dev/","source":"https://aws.amazon.com/products/"},{"id":"aws-codeguru","provider":"aws","name":"Amazon CodeGuru","category":"devtools","also":["ai-apps"],"tagline":"Automated code review and finding slow, costly lines of code.","description":"Amazon CodeGuru uses machine learning to review code and help tune it. Its Reviewer part finds defects in code, and its Profiler part locates the lines that cost the most to run. If you already have a team pushing code and want an automatic second opinion alongside human review, that is where it fits. Pricing is based on lines of code for review or sampling hours for profiling. Check AWS's current guidance before starting anything new on it.","useCases":["Review code for defects automatically","Spot expensive lines of code","Find code using the most resources"],"pricing":"Per lines of code or sampling hour.","status":"ga","url":"https://aws.amazon.com/codeguru/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-devops-guru","provider":"aws","name":"Amazon DevOps Guru","category":"observability","also":["ai-apps"],"tagline":"Spots unusual behavior in your AWS app and suggests fixes.","description":"Amazon DevOps Guru watched your app on AWS and flagged unusual behavior, using machine learning (software that learns patterns from data). If your site suddenly slowed down or started throwing errors overnight, it could point to the likely cause and suggest a fix. It learned what normal looks like for your setup, so you did not have to write alert rules by hand. AWS has announced it is retiring this service, with end of support on September 30, 2027.","useCases":["Detect unusual app behavior automatically","Get suggested fixes for problems","Review operational insights after incidents"],"pricing":"Per resource analyzed per hour.","status":"deprecated","url":"https://aws.amazon.com/devops-guru/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-healthimaging","provider":"aws","name":"AWS HealthImaging","category":"specialized","also":["object-storage"],"tagline":"Store medical scans like X-rays and MRIs in the cloud, ready to view.","description":"AWS HealthImaging stores medical images such as X-rays, CT scans and MRIs in DICOM, the standard file format hospitals use for scans. If you are building a tool where doctors review scans in a browser, or training an AI model on imaging data, it retrieves images quickly for both. It is HIPAA-eligible, which means it can be used for US patient-privacy requirements. Pick it over general file storage when scans are your main data and speed matters. You pay per GB stored and per request.","useCases":["Store medical imaging files securely","Load scans quickly in viewers","Feed medical images to AI"],"pricing":"Per GB stored and per request.","status":"ga","url":"https://aws.amazon.com/healthimaging/","source":"https://aws.amazon.com/products/"},{"id":"aws-healthlake","provider":"aws","name":"AWS HealthLake","category":"specialized","also":["warehouse"],"tagline":"Keep and search patient health records in one standard format.","description":"AWS HealthLake is a managed place to keep patient health records in FHIR, a common standard format for sharing health data, and to search them. If you are building a patient app or a dashboard that pulls records from several clinics, it tidies the data into one consistent shape so you can query it. Pick it over a general database when you want health-record standards handled for you instead of building that yourself. You pay per datastore hour, plus storage and queries.","useCases":["Store clinical records in FHIR format","Normalize and query patient records","Analyze health data across populations"],"pricing":"Per datastore hour, storage and queries.","status":"ga","url":"https://aws.amazon.com/healthlake/","source":"https://aws.amazon.com/products/"},{"id":"aws-healthomics","provider":"aws","name":"AWS HealthOmics","category":"specialized","also":["workflows"],"tagline":"Store genetic data and run the analysis steps on it.","description":"AWS HealthOmics stores genetic sequence data and runs the chains of analysis steps, called workflows, that scientists use to make sense of it. If you are building a genomics product, you can keep DNA data here and run your analysis pipeline on it without piecing together storage and computing yourself. It is built for this kind of data, which is a good reason to pick it over general-purpose tools. The variant and annotation stores are in maintenance, so check before building on them. Pricing is usage-based, per workflow run and GB stored.","useCases":["Run genomic analysis workflows","Store sequence data for research","Process omics data at scale"],"pricing":"Usage-based per workflow run and GB stored.","status":"ga","url":"https://aws.amazon.com/healthomics/","source":"https://aws.amazon.com/products/"},{"id":"aws-monitron","provider":"aws","name":"Amazon Monitron","category":"iot","tagline":"Vibration sensors that warn you before a machine breaks down.","description":"Amazon Monitron is a ready-made system that uses vibration sensors on equipment to warn you before a machine fails. If you run a workshop or factory, you attach sensors to motors and pumps and get alerts when something starts to look off. Sensors, a gateway (a small device that sends their readings to the cloud) and machine learning come together, so you do not build the pieces yourself. Pricing is per sensor and gateway. It is closed to new customers, so you cannot sign up for it now.","useCases":["Track equipment condition with sensors","Spot wear from vibration readings","Predict maintenance needs for machines"],"pricing":"Per sensor and gateway.","status":"closed","url":"https://aws.amazon.com/monitron/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-deep-learning-amis","provider":"aws","name":"AWS Deep Learning AMIs","category":"ai-platform","also":["vms"],"tagline":"Ready-made AWS servers with AI tools already installed.","description":"AWS Deep Learning AMIs are ready-to-launch server setups with popular AI tools like PyTorch and TensorFlow already installed, along with GPU drivers and CUDA (NVIDIA's software that lets graphics chips do AI work). If you want to train an AI model and would rather not spend a day installing and matching software versions, you launch one and start working. Pick it over a blank server when you want a quick GPU setup. The images are free, and you only pay for the EC2 server (AWS's rentable computer) underneath.","useCases":["Launch a GPU machine ready for training","Start with PyTorch or TensorFlow preinstalled","Skip manual driver and CUDA setup"],"pricing":"Free. Pay for EC2.","status":"ga","url":"https://aws.amazon.com/ai/machine-learning/amis/","source":"https://aws.amazon.com/products/"},{"id":"aws-deep-learning-containers","provider":"aws","name":"AWS Deep Learning Containers","category":"ai-platform","also":["containers"],"tagline":"Prebuilt software packages for training and running AI models.","description":"AWS Deep Learning Containers are prebuilt Docker images, meaning packaged bundles of software that run the same way anywhere, with AI frameworks already set up. If your team trains or serves a model and wants everyone on identical, working software, you start from these instead of building your own. They are tested to run on EKS and ECS (AWS services that run containers) and on SageMaker (AWS's service for building AI models). The images are free, and you pay only for the computing you use.","useCases":["Run ML models in standard containers","Deploy training on EKS, ECS or SageMaker","Use tested framework images for inference"],"pricing":"Free. Pay for compute.","status":"ga","url":"https://aws.amazon.com/ai/machine-learning/containers/","source":"https://aws.amazon.com/products/"},{"id":"aws-trainium","provider":"aws","name":"AWS Trainium and Inferentia","category":"gpu","also":["ai-platform"],"tagline":"AWS's own AI chips, aimed at cheaper model training and running.","description":"AWS Trainium and Inferentia are chips AWS designed for AI work: Trainium for training models and Inferentia for running them. If you are training a large language model or serving one to your users and GPU bills keep climbing, these chips offer a lower-cost option on EC2 (AWS's rentable servers) and SageMaker. You use them through the Neuron SDK, AWS's toolkit for working with the chips. Pick them over standard GPUs when cost matters most. You pay per instance-hour for Trn and Inf instances.","useCases":["Train large language models at lower cost","Run AI inference more cheaply","Use custom AWS chips via Neuron SDK"],"pricing":"Per instance-hour for Trn and Inf instances.","status":"ga","url":"https://aws.amazon.com/ai/machine-learning/trainium/","source":"https://aws.amazon.com/products/"},{"id":"aws-iam","provider":"aws","name":"AWS Identity and Access Management (IAM)","category":"identity","tagline":"Decide who can do what in your AWS account.","description":"AWS Identity and Access Management (IAM) is where you control who can do what in your AWS account. If a contractor needs to read your files but must never delete your database, you give them a role, which is a bundle of permissions, that allows only that. It supports temporary credentials and detailed conditions, which makes it easy to follow least privilege, meaning everyone gets only the access they need. It is built into AWS and free to use.","useCases":["Give each person only needed permissions","Grant roles access across accounts","Issue temporary credentials instead of permanent keys"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/iam/","source":"https://aws.amazon.com/products/"},{"id":"aws-iam-identity-center","provider":"aws","name":"AWS IAM Identity Center","category":"identity","tagline":"One login for your team across AWS accounts and work apps.","description":"AWS IAM Identity Center gives your team single sign-on, meaning one login that opens everything they are allowed to use. If your company has several AWS accounts and a few work apps, people sign in once instead of juggling separate passwords. It connects to existing sign-in providers like Okta or Entra ID, or you can use its built-in directory of users. Pick it over managing individual users in each account once your team grows past a handful. It is free.","useCases":["Sign staff into all AWS accounts once","Connect Okta or Entra ID logins","Manage access across many accounts"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/iam/identity-center/","source":"https://aws.amazon.com/products/"},{"id":"aws-cognito","provider":"aws","name":"Amazon Cognito","category":"identity","tagline":"Add customer sign-up and login to your app without building it.","description":"Amazon Cognito handles sign-up and login for your app's customers. If you want people to join with an email and password, or tap Sign in with Google, and add two-step verification (MFA) or passkeys, you can skip building all of that yourself. It can also give signed-in users temporary AWS access through identity pools. You pay per monthly active user, with a free tier of 10,000 on some plans. Cognito Sync, an older add-on, is in maintenance.","useCases":["Add sign-up and sign-in to apps","Offer social login and passkeys","Give app users temporary AWS access"],"pricing":"Per monthly active user. Free tier of 10,000 MAU for some plans.","status":"ga","url":"https://aws.amazon.com/cognito/","source":"https://aws.amazon.com/products/"},{"id":"aws-directory-service","provider":"aws","name":"AWS Directory Service","category":"identity","tagline":"Microsoft Active Directory, run for you on AWS.","description":"AWS Directory Service runs Microsoft Active Directory, the system many companies use to manage staff accounts and Windows computers, on AWS. If you are moving Windows-based apps to the cloud and need servers to join your company's domain, this gives you a managed directory without maintaining the machines yourself. It can also connect to your office's existing directory. Pick it when your setup depends on Active Directory. Simple AD, one of its options, is in maintenance. You pay per directory-hour.","useCases":["Run Microsoft Active Directory on AWS","Join Windows servers to a domain","Connect to on-premises directories"],"pricing":"Per directory-hour.","status":"ga","url":"https://aws.amazon.com/directoryservice/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloud-directory","provider":"aws","name":"Amazon Cloud Directory","category":"identity","tagline":"A legacy store for organizing data in a tree-like structure.","description":"Amazon Cloud Directory is a database built for information that naturally fits into a tree, where items sit under other items, like folders inside folders. If you already run an app that keeps this kind of hierarchical data in it, such as a custom structure of people, groups or devices, it can keep serving that role. Pricing is based on how many GB you store and how many requests you make. It is closed to new customers, so if you are starting a new project, pick a different service for storing your data.","useCases":["Keep an existing app's tree data","Keep existing hierarchical data running","Store people, groups or devices in trees"],"pricing":"Per GB and request.","status":"closed","url":"https://aws.amazon.com/cloud-directory/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-verified-permissions","provider":"aws","name":"Amazon Verified Permissions","category":"identity","tagline":"Decide who can do what inside your app, without hardcoding the rules.","description":"Amazon Verified Permissions is a managed service that keeps your app's access rules in one place and answers the question \"is this person allowed to do this?\" If your app has admins, editors and viewers, or lets users share documents with specific people, you can write those rules once instead of scattering them through your code. Rules are written in Cedar, a policy language made for permissions, and can depend on a person's role or on details like who owns a file. Choose it when your access rules are getting complicated and you want to change them without redeploying your app. You pay per authorization request.","useCases":["Write app access rules in one place","Check if a user may do something","Share documents with specific people"],"pricing":"Per authorization request.","status":"ga","url":"https://aws.amazon.com/verified-permissions/","source":"https://aws.amazon.com/products/"},{"id":"aws-verified-access","provider":"aws","name":"AWS Verified Access","category":"zero-trust","tagline":"Let your team and contractors reach internal apps without a VPN.","description":"AWS Verified Access lets people reach your private, internal apps from anywhere without installing a VPN (a tool that connects someone to a company network). Before each request goes through, it checks who the person is and whether their device meets your security rules. If you have an internal dashboard or admin tool and a contractor needs access to just that, you can grant it without opening up your whole network. It is a good pick when you want to replace a VPN with something more precise. You pay per application per hour, plus for the data processed.","useCases":["Give contractors access to one internal app","Replace a VPN for internal tools","Check device health before each request"],"pricing":"Per application-hour plus data processed.","status":"ga","url":"https://aws.amazon.com/verified-access/","source":"https://aws.amazon.com/products/"},{"id":"aws-guardduty","provider":"aws","name":"Amazon GuardDuty","category":"threat-intel","tagline":"Watches your AWS account around the clock and flags suspicious activity.","description":"Amazon GuardDuty is a security watchdog that scans the activity records of your AWS account for signs of trouble. If someone steals your login keys and starts spinning up servers, or malware lands on your files, GuardDuty can raise an alert. It looks at account activity logs, network traffic records and DNS lookups, and it also covers S3 file storage, EKS (container hosting) and running workloads. You do not have to set up your own monitoring, which makes it a sensible first security tool to switch on. It costs based on how much activity it analyzes, and there is a 30-day free trial.","useCases":["Flag stolen keys and suspicious server launches","Detect malware on files and workloads","Watch account activity around the clock"],"pricing":"Per volume of events analyzed. 30-day free trial.","status":"ga","url":"https://aws.amazon.com/guardduty/","source":"https://aws.amazon.com/products/"},{"id":"aws-inspector","provider":"aws","name":"Amazon Inspector","category":"threat-intel","tagline":"Scans your servers and code packages for known security holes.","description":"Amazon Inspector automatically checks your software for known security weaknesses, the kind that get public IDs called CVEs. It scans EC2 servers, container images (packaged apps ready to run) and Lambda functions (small pieces of code that run on demand), and it also points out network exposure. If you ship your app as a container, Inspector can check each image stored in ECR, AWS's container storage, before it goes live. It runs on its own, so you do not have to remember to run scans. You pay per instance or image scanned.","useCases":["Scan servers and functions for known flaws","Check container images before they go live","Find exposed network paths automatically"],"pricing":"Per instance or image scanned.","status":"ga","url":"https://aws.amazon.com/inspector/","source":"https://aws.amazon.com/products/"},{"id":"aws-macie","provider":"aws","name":"Amazon Macie","category":"threat-intel","tagline":"Finds personal data and passwords hiding in your S3 file storage.","description":"Amazon Macie looks through your S3 storage buckets (AWS's place for files) and tells you where sensitive information is sitting. If your app lets users upload documents, you may have names, addresses or even passwords stored in places you forgot about. Macie uses machine learning and pattern matching to spot personal data (often called PII) and credentials. It is a good fit when you need to prove to a customer or auditor that you know where private data lives. You pay per bucket monitored plus the amount of data scanned.","useCases":["Find personal data in S3 storage","Spot exposed passwords in file buckets","Prove where private data is stored"],"pricing":"Per bucket monitored plus data scanned.","status":"ga","url":"https://aws.amazon.com/macie/","source":"https://aws.amazon.com/products/"},{"id":"aws-detective","provider":"aws","name":"Amazon Detective","category":"threat-intel","tagline":"Helps you trace how a security problem started and what it touched.","description":"Amazon Detective helps you dig into a security alert to find out what really happened. It builds a connected picture of how your resources behaved over time from your logs, so you can follow the trail back to the root cause. If GuardDuty warns you about a strange login, Detective lets you see what that account did before and after. It is a good choice when an alert is not enough and you need answers, not just a warning. You pay per GB of data it takes in.","useCases":["Trace how a security problem started","See what an account did before alerts","Investigate incidents beyond basic warnings"],"pricing":"Per GB of data ingested.","status":"ga","url":"https://aws.amazon.com/detective/","source":"https://aws.amazon.com/products/"},{"id":"aws-security-hub","provider":"aws","name":"AWS Security Hub","category":"threat-intel","also":["management"],"tagline":"One dashboard for your security alerts and compliance checks.","description":"AWS Security Hub pulls security alerts from AWS tools and partner products into one place, so you do not have to check a dozen screens. It also runs automated checks against well-known rulebooks, such as CIS (a common security checklist) and PCI (the rules for handling card payments). If a customer asks whether your setup follows good security practice, this gives you a clear answer. It makes the most sense once you use more than one security tool. You pay per check and per finding it collects.","useCases":["Show security alerts from many tools together","Check setup against CIS and PCI rulebooks","Answer customer security checklist questions"],"pricing":"Per check and per finding ingested.","status":"ga","url":"https://aws.amazon.com/security-hub/","source":"https://aws.amazon.com/products/"},{"id":"aws-security-lake","provider":"aws","name":"Amazon Security Lake","category":"threat-intel","tagline":"Gather all your security logs into one searchable store.","description":"Amazon Security Lake collects security logs from AWS, software-as-a-service tools and your own systems, then stores them together in your own S3 storage. It converts them to OCSF, a shared format, so different tools can read them the same way. If you run several AWS accounts and your logs are scattered, this puts them in one spot for your security team or analysis tools (often called a SIEM). Pick it when you want to keep and search all your security data together. You pay per GB it takes in and converts.","useCases":["Store security logs from all accounts together","Convert logs into one shared format","Search security data in one place"],"pricing":"Per GB ingested and normalized.","status":"ga","url":"https://aws.amazon.com/security-lake/","source":"https://aws.amazon.com/products/"},{"id":"aws-security-incident-response","provider":"aws","name":"AWS Security Incident Response","category":"threat-intel","tagline":"Get expert help around the clock when a security incident hits.","description":"AWS Security Incident Response helps you cope when something goes wrong, such as a break-in or a ransomware attack (where criminals lock your data and demand money). It sorts through alerts automatically and gives you access to the AWS Customer Incident Response Team, real AWS security specialists. If you do not have your own security staff, it means you are not facing a crisis alone at 2 a.m. It suits small teams that cannot hire full-time experts. You pay per finding ingested, with a subscription tier.","useCases":["Triage security alerts automatically","Get help from AWS security specialists","Respond to ransomware or break-ins"],"pricing":"Per findings ingested, with a subscription tier.","status":"ga","url":"https://aws.amazon.com/security-incident-response/","source":"https://aws.amazon.com/products/"},{"id":"aws-security-agent","provider":"aws","name":"AWS Security Agent","category":"threat-intel","also":["agents"],"tagline":"An AI helper that hunts for security flaws in your app and suggests fixes.","description":"AWS Security Agent is an AI service that acts like a security reviewer for the apps you build. It looks over your designs and code, runs penetration tests (safe, simulated attacks that reveal weak spots), and suggests how to fix what it finds. If you build quickly with AI tools and have no security engineer on the team, it can catch problems before real attackers do. It is a handy pick when you want regular checks without hiring a testing firm. Pricing is based on usage.","useCases":["Review app designs and code for flaws","Run automated penetration tests","Suggest fixes for weak spots"],"pricing":"Usage-based.","status":"ga","url":"https://aws.amazon.com/security-agent/","source":"https://aws.amazon.com/products/"},{"id":"aws-waf","provider":"aws","name":"AWS WAF","category":"waf","also":["bots"],"tagline":"A shield that blocks bad web traffic before it reaches your app.","description":"AWS WAF is a web application firewall, a filter that sits in front of your website or API and blocks harmful requests. If your site has a login form or search box, attackers may try tricks like SQL injection or XSS (ways of sneaking in malicious commands). WAF can stop those with ready-made rules, limit how often one visitor can send requests, and block unwanted bots. It works with CloudFront, ALB, API Gateway and more, so you can add protection without changing your code. You pay per web ACL (a set of rules), per rule and per million requests.","useCases":["Block common hacking tricks sent through forms","Limit how often one visitor can make requests","Block unwanted bots from your site"],"pricing":"Per web ACL, rule and million requests.","status":"ga","url":"https://aws.amazon.com/waf/","source":"https://aws.amazon.com/products/"},{"id":"aws-shield","provider":"aws","name":"AWS Shield","category":"ddos","tagline":"Protection against floods of fake traffic that knock your site offline.","description":"AWS Shield protects your site or app from DDoS attacks, where attackers flood you with fake traffic to knock you offline. If your app sits behind CloudFront (AWS's content delivery service) or an ELB (a load balancer that spreads visitors across servers), Shield Standard guards it automatically at no cost. Most small teams never need more than that. Shield Advanced, at about $3,000 per month, adds faster detection, access to an AWS response team and cost protection, which suits businesses that cannot afford any downtime.","useCases":["Defend CloudFront and load balancers from floods","Keep sites up during traffic floods","Get a response team during attacks"],"pricing":"Standard free. Advanced about $3,000 per month.","status":"ga","url":"https://aws.amazon.com/shield/","source":"https://aws.amazon.com/products/"},{"id":"aws-firewall-manager","provider":"aws","name":"AWS Firewall Manager","category":"waf","also":["management"],"tagline":"Set your firewall rules once and apply them across all your AWS accounts.","description":"AWS Firewall Manager lets you set security rules in one place and apply them across every account in your company's AWS setup, which AWS calls an Organization. If you keep separate accounts for testing and your live product, you can make sure each one gets the same web firewall (WAF) and attack protection instead of configuring them by hand. It covers WAF, Shield, security groups (simple allow and block rules for servers) and Network Firewall policies. Pick it when you have many accounts and need rules enforced the same way everywhere.","useCases":["Apply firewall rules across many accounts","Set one security policy for the whole organization","Manage WAF and Shield rules centrally"],"pricing":"Per policy per Region.","status":"ga","url":"https://aws.amazon.com/firewall-manager/","source":"https://aws.amazon.com/products/"},{"id":"aws-network-firewall","provider":"aws","name":"AWS Network Firewall","category":"networking","also":["waf"],"tagline":"A managed firewall that checks traffic in and out of your private network.","description":"AWS Network Firewall is a managed firewall for your VPC, the private network where your AWS servers and databases live. It inspects traffic coming in and going out, can block known attack patterns (intrusion prevention), and can filter by domain name. A common use is egress filtering: letting your servers reach only the websites you approve, so a compromised server can't send data anywhere it likes. Pick it over basic security groups when you need deeper inspection. You pay per firewall-hour plus per GB processed.","useCases":["Filter which websites your servers can reach","Detect and block network intrusions","Inspect traffic at the edge of your network"],"pricing":"Per firewall-hour plus per GB processed.","status":"ga","url":"https://aws.amazon.com/network-firewall/","source":"https://aws.amazon.com/products/"},{"id":"aws-network-security-manager","provider":"aws","name":"AWS Network Security Manager","category":"networking","also":["waf"],"tagline":"Write network security policies and roll them out to AWS network controls.","description":"AWS Network Security Manager helps you define security policies for your networks and then push them out to the network controls AWS provides. If you have rules about which traffic is allowed where, and you want them applied consistently instead of set up one piece at a time, this is the tool for rolling them out. It is aimed at teams that manage network rules in more than one place. AWS does not publish a price for it, so you need to contact AWS for pricing.","useCases":["Write network security rules once","Push security policies to network controls","Roll out network protections across environments"],"pricing":"Contact AWS for pricing.","status":"ga","url":"https://aws.amazon.com/network-security-manager/","source":"https://aws.amazon.com/products/"},{"id":"aws-acm","provider":"aws","name":"AWS Certificate Manager (ACM)","category":"certificates","tagline":"Free HTTPS certificates for your site that renew themselves.","description":"AWS Certificate Manager gives your site the certificate that puts the padlock and https in the browser, using TLS, the standard that encrypts traffic between visitors and your app. If you serve your app through CloudFront (a content delivery service) or an ELB (a load balancer), you can get a public certificate for free, and AWS renews it automatically so it never expires on you. The catch is that free certificates work with AWS integrated services. Pick it over buying certificates elsewhere when your app already lives on AWS.","useCases":["Get free HTTPS certificates for AWS services","Renew certificates automatically without manual work","Secure sites on CloudFront and load balancers"],"pricing":"Public certificates are free.","status":"ga","url":"https://aws.amazon.com/acm/","source":"https://aws.amazon.com/products/"},{"id":"aws-private-ca","provider":"aws","name":"AWS Private Certificate Authority","category":"certificates","tagline":"Issue your own certificates for internal apps, devices and people.","description":"AWS Private Certificate Authority lets you run your own certificate issuer, a private version of the kind that secures public websites. If your company has internal tools, or devices that need to prove who they are, you can hand out certificates for encrypted internal connections (TLS) and device identity without running the underlying software yourself. This setup is often called PKI, a system for issuing and trusting certificates. Pick it instead of AWS Certificate Manager's free public certificates when what you're securing isn't a public website. You pay per authority per month plus per certificate.","useCases":["Create your own private certificate authority","Issue certificates for internal apps","Give devices their own identity certificates"],"pricing":"Per CA per month plus per certificate.","status":"ga","url":"https://aws.amazon.com/private-ca/","source":"https://aws.amazon.com/products/"},{"id":"aws-kms","provider":"aws","name":"AWS Key Management Service (KMS)","category":"secrets","tagline":"Create and control the encryption keys that lock your data on AWS.","description":"AWS Key Management Service creates and guards the encryption keys that scramble your data so only approved people and services can read it. If you store customer files in S3 (AWS's file storage) or use disks attached to servers (EBS), KMS is the usual place to hold the keys, since it works with most AWS services. Every use of a key can be audited through CloudTrail, AWS's activity log. You can also use external key stores if you want keys kept outside AWS. Pricing is per key per month plus per 10,000 requests.","useCases":["Create and manage encryption keys","Encrypt data stored in AWS services","Log every use of keys for audits"],"pricing":"Per key per month plus per 10,000 requests.","status":"ga","url":"https://aws.amazon.com/kms/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloudhsm","provider":"aws","name":"AWS CloudHSM","category":"secrets","tagline":"Your own dedicated security hardware for storing encryption keys.","description":"AWS CloudHSM gives you a security device, called a hardware security module (HSM), that is dedicated to you alone and sits in your VPC, your private network on AWS. You control the keys. Use it when rules or customers require regulated key storage, or when you are running your own certificate system. The devices meet FIPS 140-3, a US government security standard. For most apps, AWS KMS is the simpler place to start, so pick CloudHSM when you specifically need to hold the keys yourself. You pay per HSM per hour.","useCases":["Store keys in dedicated hardware you control","Meet strict rules for regulated key storage","Build a custom certificate system"],"pricing":"Per HSM-hour.","status":"ga","url":"https://aws.amazon.com/cloudhsm/","source":"https://aws.amazon.com/products/"},{"id":"aws-secrets-manager","provider":"aws","name":"AWS Secrets Manager","category":"secrets","tagline":"Keep passwords and API keys out of your code, and change them automatically.","description":"AWS Secrets Manager is a locked vault for the passwords and API keys your app needs, such as database logins. Instead of pasting them into your code or settings files, your app asks for them when it runs. Secrets Manager can also rotate them automatically, meaning it changes a password on a schedule so a leaked one stops working. Access can be set in fine detail, so each part of your app sees only what it needs. You pay per secret per month plus per API call.","useCases":["Store database passwords safely","Keep API keys out of your code","Rotate passwords automatically on a schedule"],"pricing":"Per secret per month plus API calls.","status":"ga","url":"https://aws.amazon.com/secrets-manager/","source":"https://aws.amazon.com/products/"},{"id":"aws-signer","provider":"aws","name":"AWS Signer","category":"certificates","also":["devtools"],"tagline":"Stamp your code as trusted so only approved code runs.","description":"AWS Signer puts a digital signature on your code, like a tamper-proof seal, and checks it later so only trusted code runs. If you use Lambda, AWS's service for running small pieces of code without managing servers, you can require signed code so unapproved changes can't slip into your app. It also covers firmware, the built-in software on IoT devices (connected gadgets). Pick it when you need proof that code hasn't been altered between your team and the place it runs. You pay per signing operation.","useCases":["Sign code so only trusted code runs","Sign code for Lambda functions","Sign firmware for connected devices"],"pricing":"Per signing operation.","status":"ga","url":"https://docs.aws.amazon.com/signer/","source":"https://aws.amazon.com/products/"},{"id":"aws-payment-cryptography","provider":"aws","name":"AWS Payment Cryptography","category":"secrets","also":["specialized"],"tagline":"Managed encryption for card payments and PINs.","description":"AWS Payment Cryptography handles the behind-the-scenes encryption used when cards and PINs are processed. If you are building something that issues cards or needs to translate PINs between systems, it gives you managed keys and payment HSM operations (special security hardware) for PIN, card and EMV workflows. EMV is the chip-card standard. It is built for companies working inside the payments chain, not for simply accepting card payments on a website, where a payment processor is the usual choice. You pay per key and per API call.","useCases":["Handle card and PIN encryption operations","Support card issuing workflows","Translate PIN data between systems"],"pricing":"Per key and per API call.","status":"ga","url":"https://aws.amazon.com/payment-cryptography/","source":"https://aws.amazon.com/products/"},{"id":"aws-artifact","provider":"aws","name":"AWS Artifact","category":"management","tagline":"Download AWS's compliance reports and sign agreements, free.","description":"AWS Artifact is a free portal where you download AWS's official compliance reports, such as SOC, ISO and PCI, and sign agreements such as a BAA, a contract needed when you handle US health data. If a customer or auditor asks you to prove your cloud provider meets security standards, this is where you get the paperwork. It is evidence about AWS itself, not a tool that checks your own app. Pick it when you are answering security questionnaires or preparing for an audit.","useCases":["Download security reports for auditors","Sign compliance agreements online","Gather proof for compliance audits"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/artifact/","source":"https://aws.amazon.com/products/"},{"id":"aws-audit-manager","provider":"aws","name":"AWS Audit Manager","category":"management","tagline":"Gather proof for compliance audits automatically from your AWS usage.","description":"AWS Audit Manager collects the evidence auditors ask for, such as proof that your setup follows security rules. If you are chasing a SOC 2 or HIPAA report (common security and health-data standards) to win bigger customers, it matches how you use AWS to those requirements and keeps gathering proof as you go. That saves you from taking screenshots by hand before every audit. You pay per piece of evidence collected. It is closed to new customers, so if you are just starting out, plan on a different tool.","useCases":["Collect evidence for SOC 2 audits","Map AWS usage to compliance controls","Track proof for HIPAA requirements"],"pricing":"Per evidence item.","status":"closed","url":"https://aws.amazon.com/audit-manager/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-ram","provider":"aws","name":"AWS Resource Access Manager (RAM)","category":"management","tagline":"Share your AWS resources with other accounts instead of copying them.","description":"AWS Resource Access Manager, or RAM, lets one AWS account share things it owns with other accounts. If your company runs separate accounts for testing and for your live product, you can share one section of a network (a subnet) so both use it instead of building two. It can also share Transit Gateways (a central network hub) and license configurations. The big draw is that you avoid duplicate setups to maintain, and the service itself is free.","useCases":["Share subnets across AWS accounts","Share Transit Gateways without copying them","Share license settings with other accounts"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/ram/","source":"https://aws.amazon.com/products/"},{"id":"aws-wickr","provider":"aws","name":"AWS Wickr","category":"messaging","also":["end-user"],"tagline":"Private team chat, calls and file sharing that only your team can read.","description":"AWS Wickr is a messaging app for teams that locks conversations, calls and files so only the people in them can read them. This is called end-to-end encryption. If your team talks about sensitive things, like legal matters or customer data, you can use it instead of a regular chat app. Admins get controls over who can do what, plus choices for how long data is kept, and it is used for government communications too. You pay per user each month.","useCases":["Run end-to-end encrypted team chat","Make secure calls and share files","Handle sensitive government communications"],"pricing":"Per user per month.","status":"ga","url":"https://aws.amazon.com/wickr/","source":"https://aws.amazon.com/products/"},{"id":"aws-vpc","provider":"aws","name":"Amazon Virtual Private Cloud (VPC)","category":"networking","tagline":"Your own private, walled-off network inside AWS.","description":"Amazon Virtual Private Cloud, or VPC, is your own private network inside AWS where your servers and databases live. If your app has a public website and a database that should never be reachable from the internet, you put the site in a public subnet (an open section) and the database in a private subnet (a closed section). You choose the address ranges and set rules for what can talk to what. VPC itself is free, but extras like NAT gateways (which let private resources reach out to the internet) and data transfer are billed.","useCases":["Isolate resources in private networks","Split networks into public and private parts","Control traffic with security groups"],"pricing":"VPC is free. NAT gateways, endpoints and transfer are billed.","status":"ga","url":"https://aws.amazon.com/vpc/","source":"https://aws.amazon.com/products/"},{"id":"aws-vpn","provider":"aws","name":"AWS Virtual Private Network (VPN)","category":"networking","tagline":"Encrypted tunnels between your office or laptop and your AWS network.","description":"AWS Virtual Private Network, or VPN, creates a scrambled, private connection over the regular internet. If your team has an office or on-site equipment that needs to talk to your AWS setup, it builds IPsec tunnels (a standard way to encrypt traffic) between the two. It also offers a managed client service based on OpenVPN, so remote workers can connect safely from their own laptops. Pick it when you want a quick link without installing dedicated cables. You pay per connection hour plus data transfer.","useCases":["Connect your office network to AWS","Give remote workers secure access","Link on-premises networks securely"],"pricing":"Per connection-hour plus data transfer.","status":"ga","url":"https://aws.amazon.com/vpn/","source":"https://aws.amazon.com/products/"},{"id":"aws-direct-connect","provider":"aws","name":"AWS Direct Connect","category":"networking","tagline":"A dedicated physical cable from your site to AWS, not the public internet.","description":"AWS Direct Connect is a private fiber line running from your office or data center to AWS. If you need to move huge amounts of data, such as years of video or records, it gives steadier speeds than the public internet and lowers the cost of data leaving AWS. Choose it over a VPN (an encrypted tunnel over the internet) when consistent performance matters. You pay per port hour plus data transferred out. Most small apps will not need it.","useCases":["Connect offices to AWS with private fiber","Move large data sets more cheaply","Keep steady speeds for hybrid networks"],"pricing":"Per port-hour plus data transfer out.","status":"ga","url":"https://aws.amazon.com/directconnect/","source":"https://aws.amazon.com/products/"},{"id":"aws-transit-gateway","provider":"aws","name":"AWS Transit Gateway","category":"networking","tagline":"One central hub that connects all your AWS networks and office networks.","description":"AWS Transit Gateway is a central router that links your AWS private networks (called VPCs) and your own on-site networks. If you have grown to many separate networks that each need to talk to the others, connecting every pair gets messy fast. Transit Gateway replaces all those individual links with a single hub that scales to thousands of connections. Pick it when your network layout starts looking like a tangle. You pay per connection hour plus per gigabyte moved.","useCases":["Connect many VPCs through one hub","Connect on-premises networks to many VPCs","Replace many peering links with one router"],"pricing":"Per attachment-hour plus per GB.","status":"ga","url":"https://aws.amazon.com/transit-gateway/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloud-wan","provider":"aws","name":"AWS Cloud WAN","category":"networking","tagline":"Build and manage a worldwide private network across AWS and your offices.","description":"AWS Cloud WAN builds and runs a private network that spans the world, connecting AWS Regions (its data center areas) and your own sites. If your company has offices and apps in several countries, you can set central rules once, such as keeping the finance network separate from the guest network, instead of configuring each location. It is a managed alternative to wiring everything together yourself. You pay per core network edge hour (each network entry point) plus each connection.","useCases":["Run a global private network","Separate networks into secure segments","Manage networks across regions centrally"],"pricing":"Per core network edge-hour and attachment.","status":"ga","url":"https://aws.amazon.com/cloud-wan/","source":"https://aws.amazon.com/products/"},{"id":"aws-privatelink","provider":"aws","name":"AWS PrivateLink","category":"networking","tagline":"Reach services privately, with no traffic crossing the public internet.","description":"AWS PrivateLink lets your private network reach a service without sending traffic over the public internet. If your app stores files in S3 (AWS file storage) or uses a SaaS tool (software you rent online), you can connect through an interface endpoint, which is a private doorway inside your network. Nothing needs a public IP address, so less is exposed to attackers. Choose it when security rules say traffic must stay on the AWS network. You pay per endpoint hour plus per gigabyte.","useCases":["Reach S3 or SaaS without public internet","Keep traffic inside the AWS network","Expose your service to other VPCs"],"pricing":"Per endpoint-hour plus per GB.","status":"ga","url":"https://aws.amazon.com/privatelink/","source":"https://aws.amazon.com/products/"},{"id":"aws-interconnect","provider":"aws","name":"AWS Interconnect","category":"networking","tagline":"Private links between AWS and other cloud companies, set up for you.","description":"AWS Interconnect gives you managed private connections between AWS and other cloud providers. If part of your product runs on AWS and another part runs on a different cloud, it lets the two talk privately instead of you building a network between them. You skip the work of setting up and maintaining that link yourself. Pick it when you are running across more than one cloud and want the connection handled for you. Pricing is based on usage.","useCases":["Connect AWS to other cloud providers","Build multicloud networks without custom links","Pay only for what you use"],"pricing":"Usage-based.","status":"ga","url":"https://aws.amazon.com/interconnect/","source":"https://aws.amazon.com/products/"},{"id":"aws-global-accelerator","provider":"aws","name":"AWS Global Accelerator","category":"networking","also":["load-balancing"],"tagline":"Fixed addresses that send users to the nearest healthy copy of your app.","description":"AWS Global Accelerator gives your app a few fixed internet addresses and sends each visitor to the closest copy of your app that is working. If you run your app in several Regions (AWS locations around the world) and one goes down, traffic shifts to a healthy one. It also improves speed for apps that use TCP and UDP (the basic ways data travels online), like online games, by using the AWS backbone, its own fast network. You pay per accelerator hour plus premium data transfer.","useCases":["Send users to the closest healthy server","Fail over to another region automatically","Speed up real-time apps like games"],"pricing":"Per accelerator-hour plus premium data transfer.","status":"ga","url":"https://aws.amazon.com/global-accelerator/","source":"https://aws.amazon.com/products/"},{"id":"aws-elb","provider":"aws","name":"Elastic Load Balancing","category":"load-balancing","tagline":"Spread incoming traffic across several servers so none gets swamped.","description":"Elastic Load Balancing splits incoming traffic across several servers so your app stays fast and keeps running if one server fails. If your website suddenly gets popular, a load balancer sends each visitor to a server that can handle them. It comes in four types: Application (for web traffic), Network (for fast, low-level traffic like TCP), Gateway (for security appliances) and Classic (the older kind). Choose it instead of pointing all visitors at one server. You pay per load balancer hour plus capacity units used.","useCases":["Spread visitor traffic across servers","Route web requests to the right service","Balance network connections across servers"],"pricing":"Per load balancer-hour plus capacity units.","status":"ga","url":"https://aws.amazon.com/elasticloadbalancing/","source":"https://aws.amazon.com/products/"},{"id":"aws-route53","provider":"aws","name":"Amazon Route 53","category":"dns","also":["domains"],"tagline":"Point your domain name at your app, and buy domains too.","description":"Amazon Route 53 connects your domain name, like yourapp.com, to the place your app actually lives. If you are launching a product, you can buy the domain here and use DNS, the internet's address book, to send visitors to your site. It can also check that your app is healthy and steer people to a backup if something breaks, or to the closest location for faster loading. Pick it when your app already runs on AWS and you want your domain and traffic routing in one place. You pay per domain you manage each month, plus per million lookups.","useCases":["Point a domain name at your app","Send users to a backup site if one fails","Register a new domain name"],"pricing":"Per hosted zone per month plus per million queries.","status":"ga","url":"https://aws.amazon.com/route53/","source":"https://aws.amazon.com/products/"},{"id":"aws-arc","provider":"aws","name":"Amazon Application Recovery Controller (ARC)","category":"networking","also":["load-balancing"],"tagline":"Move your app's traffic away from a broken zone or region.","description":"Amazon Application Recovery Controller helps you move traffic away from trouble when part of AWS has a problem. If your app serves paying customers and has to stay online, you can use it to shift users from one Region (a geographic cluster of AWS data centers) to another, or away from a single zone (one data center within a Region). Pick it when your app already runs in more than one place and you want a dependable switch to flip during an outage. Pricing is per cluster-hour for routing control. The Readiness Check feature is in maintenance.","useCases":["Switch traffic between regions in a disaster","Move traffic away from one zone quickly","Control which regions receive user requests"],"pricing":"Per cluster-hour for routing control.","status":"ga","url":"https://aws.amazon.com/application-recovery-controller/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloudfront","provider":"aws","name":"Amazon CloudFront","category":"cdn","also":["edge-compute"],"tagline":"Make your site and videos load fast for visitors anywhere.","description":"Amazon CloudFront keeps copies of your website, images and videos in hundreds of places around the world so they load fast for nearby visitors. This is called a CDN, short for content delivery network. If your product has customers in several countries, you put CloudFront in front of your site so someone in Tokyo does not wait on a server in Virginia. Pick it when speed matters for a global audience or you stream video. It has 600+ locations, lets you run small bits of code close to visitors, and includes 1 TB of free traffic out per month.","useCases":["Speed up pages for visitors worldwide","Stream videos to viewers across the globe","Run small code near visitors"],"pricing":"Per GB out and per request. Free tier: 1 TB per month.","status":"ga","url":"https://aws.amazon.com/cloudfront/","source":"https://aws.amazon.com/products/"},{"id":"aws-rtb-fabric","provider":"aws","name":"AWS RTB Fabric","category":"specialized","also":["networking"],"tagline":"Fast private connections between online ad buyers and sellers.","description":"AWS RTB Fabric is a specialized network for advertising companies. Real-time bidding is the auction that happens in the instant a page loads, deciding which ad appears. If you are building an ad exchange, or a tool that buys ad space (a DSP) or sells it (an SSP), you can use RTB Fabric to link up with partners over private, low-latency connections, meaning very short delays. Pick it when your business lives in that auction and every fraction of a second counts. It is not meant for general apps. Pricing is based on usage.","useCases":["Connect ad exchanges with buyers and sellers","Link ad platforms with low delay connections","Build a real-time ad bidding network"],"pricing":"Usage-based.","status":"ga","url":"https://aws.amazon.com/rtb-fabric/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloudwatch","provider":"aws","name":"Amazon CloudWatch","category":"observability","tagline":"See how your app is doing with charts, logs and alerts.","description":"Amazon CloudWatch is the monitoring service built into AWS. If your app slows down or crashes at 2 a.m., you can set an alarm that notifies you, search your logs (the running diary your app writes), and watch dashboards of key numbers. It can also test your site on a schedule and show how real visitors experience it. Pick it as your first monitoring tool if you are on AWS, since it is already built in. You pay per metric, per GB of logs and per alarm, and a free tier is included.","useCases":["Set alarms when something breaks","Search application logs for errors","Build dashboards for key numbers"],"pricing":"Per metric, GB of logs and alarm. Free tier included.","status":"ga","url":"https://aws.amazon.com/cloudwatch/","source":"https://aws.amazon.com/products/"},{"id":"aws-xray","provider":"aws","name":"AWS X-Ray","category":"observability","tagline":"Follow a request through your app to find what is slow or broken.","description":"AWS X-Ray follows a single user action as it passes through the different parts of your app, so you can see where it slows down or fails. If your app is split into many small services (often called microservices) and checkout feels sluggish, X-Ray draws a map that shows which piece is the culprit. Pick it over digging through logs when the problem could be hiding in any of several places. It supports OpenTelemetry, a widely used open standard for collecting this data. You pay per trace recorded and retrieved.","useCases":["Trace a slow request across services","Map how your services connect","Find where errors start"],"pricing":"Per trace recorded and retrieved.","status":"ga","url":"https://aws.amazon.com/xray/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloudtrail","provider":"aws","name":"AWS CloudTrail","category":"management","also":["observability"],"tagline":"A record of who did what in your AWS account.","description":"AWS CloudTrail keeps a log of actions taken in your AWS account, such as who created a server or changed a setting. If something unexpected happens, like a deleted file or a surprise bill, you can check it to see who did it and when. It is also the record auditors and security teams ask for. Pick it as your baseline for accountability, since one copy of management events (records of account-level changes) is free. CloudTrail Lake, its feature for searching those records, is closed to new customers.","useCases":["Record who changed what in AWS","Keep an audit trail for reviews","Investigate suspicious account activity"],"pricing":"One management event copy free. Extra events billed.","status":"ga","url":"https://aws.amazon.com/cloudtrail/","source":"https://aws.amazon.com/products/"},{"id":"aws-config","provider":"aws","name":"AWS Config","category":"management","tagline":"Track how your AWS setup changes and check it against your rules.","description":"AWS Config keeps a history of how your AWS resources are set up and flags anything that breaks your rules. If a teammate accidentally makes a storage bucket (a cloud folder for files) public, Config can catch it and show what changed. Pick it alongside CloudTrail: CloudTrail tells you who did something, while Config shows what your setup looked like before and after. It suits teams that need to prove they follow security or compliance requirements. You pay per configuration item recorded and per rule check.","useCases":["Track changes to cloud settings over time","Check resources against compliance rules","Flag settings that drift from policy"],"pricing":"Per configuration item and rule evaluation.","status":"ga","url":"https://aws.amazon.com/config/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloudformation","provider":"aws","name":"AWS CloudFormation","category":"devtools","also":["management"],"tagline":"Describe your whole AWS setup in a file and rebuild it anytime.","description":"AWS CloudFormation lets you write down your servers, databases and settings in a text file, in JSON or YAML formats, and AWS builds them for you. This approach is called infrastructure as code. If you want a test version of your app that matches the live one, you reuse the same file instead of clicking through menus again. It can also spot drift, meaning someone changed things by hand, and roll out to many accounts at once. It costs nothing for AWS resource types.","useCases":["Rebuild the same setup in new environments","Roll out one setup across many accounts","Detect settings changed by hand"],"pricing":"Free for AWS resource types.","status":"ga","url":"https://aws.amazon.com/cloudformation/","source":"https://aws.amazon.com/products/"},{"id":"aws-cdk","provider":"aws","name":"AWS Cloud Development Kit (AWS CDK)","category":"devtools","tagline":"Write your AWS setup in a regular programming language.","description":"The AWS Cloud Development Kit, or CDK, lets you describe your cloud setup in TypeScript, Python, Java and other languages instead of a long settings file. It then converts your code into CloudFormation, the AWS template system that does the actual building. If you already write app code, you can create a database and a server in a few lines and reuse building blocks, called constructs, across projects. Pick it over raw CloudFormation if you prefer code with type checking. The CDK is free.","useCases":["Describe cloud setup in regular code","Reuse building blocks across projects","Generate CloudFormation templates from code"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/cdk/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloud-control-api","provider":"aws","name":"AWS Cloud Control API","category":"devtools","tagline":"One consistent way to manage many kinds of cloud resources.","description":"AWS Cloud Control API gives you one standard way to create, read, update, delete and list hundreds of resource types, including some from third parties. If you are building a tool that manages many kinds of cloud pieces, you learn one pattern instead of a different one for each service. Pick it when you are making tooling like that. If you just want to launch an app, CloudFormation or the CDK are usually the friendlier place to start. Cloud Control API is free.","useCases":["Manage many cloud resource types one way","List resources across many service types","Manage some third-party resources too"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/cloudcontrolapi/","source":"https://aws.amazon.com/products/"},{"id":"aws-infrastructure-composer","provider":"aws","name":"AWS Infrastructure Composer","category":"devtools","tagline":"Drag and drop to design your AWS setup visually.","description":"AWS Infrastructure Composer is a visual designer where you drag pieces onto a canvas to plan how your AWS resources fit together. If you are new to serverless apps (ones where AWS runs the servers for you), it lets you see the whole picture and builds the CloudFormation template behind it. Pick it when you think better in diagrams than in code. It is free. The standalone console reaches end of support on December 7, 2026.","useCases":["Design cloud setups by dragging pieces","Build serverless templates visually","Sketch templates without writing code"],"pricing":"Free.","status":"deprecated","url":"https://aws.amazon.com/infrastructure-composer/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-cloudshell","provider":"aws","name":"AWS CloudShell","category":"devtools","tagline":"A free command line in your browser, already signed in to AWS.","description":"AWS CloudShell is a command line (a text window where you type instructions to a computer) that opens inside your browser, already signed in with your AWS account. If a tutorial or AI tool tells you to run a quick AWS command, you can paste it here instead of installing anything on your laptop. It comes with the AWS CLI, AWS's own command line tool, already set up, plus 1 GB of storage for your files. It costs nothing, so it is a good pick for quick one-off jobs.","useCases":["Run quick cloud commands from the browser","Test command-line tasks without installing tools","Keep 1 GB of files between sessions"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/cloudshell/","source":"https://aws.amazon.com/products/"},{"id":"aws-cloud9","provider":"aws","name":"AWS Cloud9","category":"devtools","tagline":"A code editor in your browser, now closed to new customers.","description":"AWS Cloud9 is a code editor (often called an IDE) that runs in your web browser, so you can write and run code without setting up your own computer. If you already have environments running in it, you can keep using them, and you pay only for the server (an EC2 instance) behind each one. It suits people who like having a ready-made coding workspace in the cloud. AWS closed it to new customers in July 2024, so anyone starting fresh should choose a different editor.","useCases":["Code in a browser-based environment","Maintain existing cloud IDE environments","Run an IDE on a cloud server"],"pricing":"Pay for the EC2 instance.","status":"closed","url":"https://aws.amazon.com/cloud9/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-codecommit","provider":"aws","name":"AWS CodeCommit","category":"devtools","tagline":"Private storage for your code, kept inside your AWS account.","description":"AWS CodeCommit is a place to keep your code and its change history privately, using Git, the standard tool for tracking versions of code. If the rest of your app already runs on AWS, it lets you keep your code in the same account instead of using a separate service. It is billed per active user each month, with a free tier for 5 users. AWS closed it to new customers in July 2024, then reversed that decision in late 2025, and it is open again with ongoing investment.","useCases":["Host private Git code repositories","Keep source code inside AWS accounts","Store code free for up to five users"],"pricing":"Per active user per month. Free tier for 5 users.","status":"ga","url":"https://aws.amazon.com/codecommit/","source":"https://aws.amazon.com/products/"},{"id":"aws-codebuild","provider":"aws","name":"AWS CodeBuild","category":"devtools","tagline":"Automatically build and test your code on machines AWS runs for you.","description":"AWS CodeBuild turns your code into something ready to ship and runs your tests, on machines AWS manages so you don't have to. If you want every code change checked automatically before it goes live, this is the step that does the building and testing. It can also run GitHub Actions jobs (GitHub's automation tool), which makes it a stand-in for runners you would otherwise host yourself. You pay per minute of build time, and the price depends on the machine size you choose.","useCases":["Compile code and run automated tests","Build app packages on managed machines","Replace self-hosted GitHub Actions runners"],"pricing":"Per build minute by instance size.","status":"ga","url":"https://aws.amazon.com/codebuild/","source":"https://aws.amazon.com/products/"},{"id":"aws-codedeploy","provider":"aws","name":"AWS CodeDeploy","category":"devtools","tagline":"Push new app versions to your servers in a controlled, gradual way.","description":"AWS CodeDeploy puts a new version of your app onto your servers and switches it on. It works with EC2 (regular AWS servers), Lambda (code that runs without a server for you to manage), ECS (AWS's container service) and servers you run yourself. If you want a bad release to reach only a few users, it supports rolling updates, and blue-green deploys, where the new version runs beside the old one before traffic moves over. It is free for EC2 and Lambda, and charges per update for your own servers.","useCases":["Roll out updates gradually across servers","Switch traffic to new versions safely","Deploy to servers, containers and functions"],"pricing":"Free for EC2 and Lambda. Per update for on-premises.","status":"ga","url":"https://aws.amazon.com/codedeploy/","source":"https://aws.amazon.com/products/"},{"id":"aws-codepipeline","provider":"aws","name":"AWS CodePipeline","category":"devtools","tagline":"Map out the steps from code change to live app and run them automatically.","description":"AWS CodePipeline lays out your release process as a series of steps, from new code arriving to the app going live, and runs them for you. If you want every change pushed to GitHub to be built, tested and deployed without clicking through each stage, you would set that up here. It connects with GitHub, CodeBuild and CloudFormation (AWS's tool for creating resources from a written file). It fits best when the rest of your stack is already on AWS. You pay per active pipeline each month.","useCases":["Automate steps from code to release","Connect GitHub, builds and deployments","Set up repeatable release pipelines"],"pricing":"Per active pipeline per month.","status":"ga","url":"https://aws.amazon.com/codepipeline/","source":"https://aws.amazon.com/products/"},{"id":"aws-codeartifact","provider":"aws","name":"AWS CodeArtifact","category":"devtools","tagline":"A private shelf for the code packages your app depends on.","description":"AWS CodeArtifact stores the packages your code is built from, including npm, PyPI and Maven ones (shared libraries for JavaScript, Python and Java). If your team writes its own reusable code and wants to share it without making it public, you can host it here as a private package registry. It can also sit between you and public sources (upstream proxying), fetching packages you use and keeping copies in your own account. You pay per GB stored and per request.","useCases":["Host private software packages for your team","Store npm and Python packages privately","Cache packages from public sources"],"pricing":"Per GB stored and request.","status":"ga","url":"https://aws.amazon.com/codeartifact/","source":"https://aws.amazon.com/products/"},{"id":"aws-codecatalyst","provider":"aws","name":"Amazon CodeCatalyst","category":"devtools","tagline":"An all-in-one place for code, builds and tasks. Closed to new customers.","description":"Amazon CodeCatalyst bundled code storage, automatic builds and issue tracking (a to-do list for bugs and features) into one service. Teams used it to keep a project's code, testing and planning together instead of spread across separate tools. It was priced per user plus the computing power used. It is closed to new customers, so it only makes sense if you already have a project running in it.","useCases":["Manage existing CodeCatalyst projects","Keep code, builds and issues together","Maintain current team development workflows"],"pricing":"Per user and compute usage.","status":"closed","url":"https://codecatalyst.aws/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-fis","provider":"aws","name":"AWS Fault Injection Service","category":"devtools","also":["observability"],"tagline":"Break things on purpose to see if your app survives.","description":"AWS Fault Injection Service lets you cause controlled failures, like stopping a server or slowing down the network, to see how your app copes. If you want to know whether your site stays up when part of it goes down, you can find out in a planned test rather than during a real outage. This practice is often called chaos engineering, and teams also use it for game days, which are practice drills for handling outages. You pay per action-minute, meaning each minute a test action runs.","useCases":["Test how systems handle failures","Stop servers on purpose to check recovery","Simulate slow networks for resilience drills"],"pricing":"Per action-minute.","status":"ga","url":"https://aws.amazon.com/fis/","source":"https://aws.amazon.com/products/"},{"id":"aws-device-farm","provider":"aws","name":"AWS Device Farm","category":"devtools","tagline":"Test your app on real phones and browsers without owning them.","description":"AWS Device Farm lets you try your app on real phones and web browsers running in the cloud, so you don't need a drawer full of devices. If you are building a mobile app, you can run automated tests across many phone models, or tap through it by hand on one. It also checks that your website works across different browsers. You pay per device-minute, or choose an unmetered plan, which is not counted by the minute, if you test a lot.","useCases":["Test mobile apps on real phones","Check websites across many browsers","Run automated tests in the cloud"],"pricing":"Per device-minute or unmetered plan.","status":"ga","url":"https://aws.amazon.com/device-farm/","source":"https://aws.amazon.com/products/"},{"id":"aws-location-service","provider":"aws","name":"Amazon Location Service","category":"specialized","tagline":"Add maps, address search, routes and tracking to your app.","description":"Amazon Location Service gives your app map features: showing maps, turning addresses into points on a map (geocoding), planning routes and tracking where things are. If you are building a delivery app, you could show a live map and trigger an alert when a driver enters an area, which is called geofencing. The map data comes from providers such as Esri, HERE and Grab. It is a handy pick when your app already runs on AWS, and you pay per request.","useCases":["Show maps inside your app","Trigger alerts when devices enter areas","Track vehicles or equipment positions"],"pricing":"Per request.","status":"ga","url":"https://aws.amazon.com/location/","source":"https://aws.amazon.com/products/"},{"id":"aws-appconfig","provider":"aws","name":"AWS AppConfig","category":"feature-flags","also":["management"],"tagline":"Turn features on or off and change settings without redeploying your app.","description":"AWS AppConfig lets you change your app's settings and switch features on or off while it is running, without publishing a new version. If you want to show a new checkout page to 10 percent of users first, you can roll it out in stages. It checks that a change is valid before it goes out and can roll it back if something goes wrong. These on and off switches are known as feature flags. You pay per configuration request and receive.","useCases":["Turn features on for some users","Roll out settings changes gradually","Undo risky configuration changes quickly"],"pricing":"Per configuration request and receive.","status":"ga","url":"https://aws.amazon.com/systems-manager/features/appconfig/","source":"https://aws.amazon.com/products/"},{"id":"aws-organizations","provider":"aws","name":"AWS Organizations","category":"management","tagline":"Manage many AWS accounts together, with one bill and shared rules.","description":"AWS Organizations lets you group several AWS accounts under one roof, so you can manage them together and pay one combined bill. If you run a product with separate accounts for testing and for live customers, you can keep them apart so a mistake in one cannot touch the other. You can also set service control policies, which are rules that limit what any account in a group is allowed to do. It costs nothing, so it is a natural first step once you outgrow a single account.","useCases":["Group many cloud accounts together","Set rules that limit account actions","Combine billing across accounts"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/organizations/","source":"https://aws.amazon.com/products/"},{"id":"aws-control-tower","provider":"aws","name":"AWS Control Tower","category":"management","tagline":"Set up a safe, organized home for many AWS accounts, rules included.","description":"AWS Control Tower sets up a ready-made, well-organized home for several AWS accounts, with safety rules already switched on. If your team is growing and you want each project or environment in its own account, it can create new accounts that follow the same standards every time. It builds on AWS Organizations and adds the setup work and guardrails (rules that stop risky changes), so you do not assemble them by hand. There is no charge for Control Tower itself, though you pay for the underlying services it uses.","useCases":["Set up a multi-account company environment","Enforce security rules across all accounts","Create new accounts with standard settings"],"pricing":"No charge. Pay for underlying services.","status":"ga","url":"https://aws.amazon.com/controltower/","source":"https://aws.amazon.com/products/"},{"id":"aws-systems-manager","provider":"aws","name":"AWS Systems Manager","category":"management","tagline":"Keep your servers patched and your settings tidy from one place.","description":"AWS Systems Manager is a toolbox for looking after your servers, whether they run on AWS or elsewhere. If you run a handful of servers and need to install security updates on all of them, you can do it in one go instead of logging into each one. It also includes Parameter Store, a safe place to keep settings and secrets your app needs, and Session Manager for connecting to a server from your browser. Many features are free and some are billed per use. Its Incident Manager and Change Manager features are closed to new customers.","useCases":["Apply software patches to many servers","Run commands on servers remotely","Store configuration values in one place"],"pricing":"Many features free. Some billed per use.","status":"ga","url":"https://aws.amazon.com/systems-manager/","source":"https://aws.amazon.com/products/"},{"id":"aws-incident-manager","provider":"aws","name":"AWS Systems Manager Incident Manager","category":"observability","also":["management"],"tagline":"Alert the right person and guide the fix when something breaks.","description":"AWS Systems Manager Incident Manager helps a team respond when something goes wrong, by alerting the right people and walking them through a fix. If an alarm in CloudWatch, AWS's monitoring service, shows your app is down, it can escalate to the on-call person and bring up a runbook, which is a step-by-step checklist for fixing the problem. It is billed per incident and per contact. It is closed to new customers, so it is only available to teams already using it.","useCases":["Run existing incident response workflows","Escalate CloudWatch alarms to contacts","Follow saved runbooks during outages"],"pricing":"Per incident and contact.","status":"closed","url":"https://docs.aws.amazon.com/incident-manager/latest/userguide/what-is-incident-manager.html","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-health","provider":"aws","name":"AWS Health","category":"management","also":["observability"],"tagline":"Get warned when an AWS problem or planned change affects your app.","description":"AWS Health tells you about AWS problems and planned changes that affect the things you actually run, rather than every outage everywhere. If your app suddenly slows down, you can check the Health Dashboard to see whether AWS is having an issue or whether the fault is on your side. You can also send alerts into other tools through EventBridge, a service that routes events to wherever you want them. The service is free, though using it through code (the API) needs a Business support plan or higher.","useCases":["Get alerts about service outages","Learn about scheduled maintenance ahead of time","Send health events into automated workflows"],"pricing":"Free. API requires Business support or higher.","status":"ga","url":"https://aws.amazon.com/health/","source":"https://aws.amazon.com/products/"},{"id":"aws-trusted-advisor","provider":"aws","name":"AWS Trusted Advisor","category":"management","tagline":"A checklist that spots wasted spend, security gaps and slow setups.","description":"AWS Trusted Advisor looks over your AWS account and gives you recommendations on cost, security and performance. If your bill has crept up and you are not sure why, it can point to things you are paying for but barely using. A small set of core checks is free for everyone. The full set needs a paid support plan, specifically Business support or higher. It is a handy quick health check before you hire an expert or dig through the account yourself.","useCases":["Check your account for cost savings","Spot security gaps and weak settings","Review performance best-practice recommendations"],"pricing":"Core checks free. Full checks with Business support.","status":"ga","url":"https://aws.amazon.com/premiumsupport/technology/trusted-advisor/","source":"https://aws.amazon.com/products/"},{"id":"aws-compute-optimizer","provider":"aws","name":"AWS Compute Optimizer","category":"management","tagline":"Learns how much your servers use and suggests better, cheaper sizes.","description":"AWS Compute Optimizer studies how much your servers and functions actually use, then suggests settings that fit better. If you picked a big server at launch to be safe and it sits mostly idle, it can tell you a smaller one would do the job for less. It covers EC2 (virtual servers), EBS (server disks), Lambda (code that runs on demand) and ECS (a service for running containers). The default tier is free, which makes it an easy way to trim your bill without guesswork.","useCases":["Find servers sized too large","Cut costs by right-sizing instances","Adjust Lambda and container settings"],"pricing":"Free for the default tier.","status":"ga","url":"https://aws.amazon.com/compute-optimizer/","source":"https://aws.amazon.com/products/"},{"id":"aws-billing","provider":"aws","name":"AWS Billing and Cost Management","category":"management","tagline":"See what you spend on AWS, set budgets and catch surprises early.","description":"AWS Billing and Cost Management is where you see your AWS bills and understand where the money goes. If you worry about a surprise bill, you can set a budget that warns you when spending passes a number you choose, and use anomaly detection to flag unusual jumps. Cost Explorer lets you chart spending over time, and cost allocation tags let you label resources so you can split costs by project or customer. Most of it is free, so it is worth turning on early.","useCases":["Set budgets and get spending alerts","Break down costs by tagged project","Spot unusual spending jumps"],"pricing":"Mostly free.","status":"ga","url":"https://aws.amazon.com/aws-cost-management/","source":"https://aws.amazon.com/products/"},{"id":"aws-savings-plans","provider":"aws","name":"Savings Plans","category":"management","also":["vms"],"tagline":"Get a lower price on AWS servers by promising to spend for 1 or 3 years.","description":"Savings Plans give you a lower price on AWS computing in return for promising to spend a set amount for one or three years. If your product has steady traffic and you know you will be running servers for a long time, the discount can reach up to 72 percent. They apply across EC2 (virtual servers), Fargate (a way to run containers without managing servers) and Lambda (code that runs on demand). Pick them over the normal pay-as-you-go rate once your usage is predictable, since you owe the commitment even if you use less.","useCases":["Lower bills for steady compute usage","Commit to spend for discounts","Cover EC2, Fargate and Lambda"],"pricing":"Discount up to 72 percent for commitment.","status":"ga","url":"https://aws.amazon.com/savingsplans/","source":"https://aws.amazon.com/products/"},{"id":"aws-flat-rate-plans","provider":"aws","name":"AWS Flat-Rate Plans","category":"management","also":["cdn"],"tagline":"One fixed monthly price for AWS web delivery and protection tools.","description":"AWS Flat-Rate Plans bundle several AWS services into one fixed monthly price, so you know the bill in advance. The bundle includes services like CloudFront (which speeds up your website for visitors worldwide), WAF (a filter that blocks bad web traffic) and DDoS protection (defense against floods of fake traffic). If a traffic spike or attack would otherwise make you nervous about a huge bill, there is no overage charge. Choose it over pay-as-you-go when a predictable cost matters most to you.","useCases":["Pay one fixed monthly price","Bundle CDN, firewall and DDoS protection","Avoid extra charges beyond the plan"],"pricing":"Fixed monthly tiers.","status":"ga","url":"https://docs.aws.amazon.com/AmazonCloudFront/latest/DeveloperGuide/flat-rate-pricing-plan.html","source":"https://aws.amazon.com/products/"},{"id":"aws-pricing-calculator","provider":"aws","name":"AWS Pricing Calculator","category":"management","tagline":"Estimate what your AWS setup will cost before you build it.","description":"AWS Pricing Calculator lets you build a cost estimate for an AWS setup before you launch anything. If you are planning an app and need a monthly number for a budget or an investor update, you can add the services you expect to use and see a projected bill. It is a better first stop than guessing from separate price pages, because it adds everything together in one place. It is free to use.","useCases":["Estimate cost before building anything","Compare architecture options by price","Plan budgets for new projects"],"pricing":"Free.","status":"ga","url":"https://calculator.aws/","source":"https://aws.amazon.com/products/"},{"id":"aws-license-manager","provider":"aws","name":"AWS License Manager","category":"management","tagline":"Keep track of your software licenses and stay within their rules.","description":"AWS License Manager helps you keep track of the software licenses you pay for, so you stay within what you are allowed to use. If your company runs software like Windows or Oracle, both on AWS and on its own machines, it can enforce your license rules so you do not go over. It matters most for business software with strict licensing terms, which can lead to costly compliance trouble if ignored. The service itself is free.","useCases":["Track software licenses across servers","Enforce Windows and Oracle license rules","Stay compliant with license terms"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/license-manager/","source":"https://aws.amazon.com/products/"},{"id":"aws-service-catalog","provider":"aws","name":"AWS Service Catalog","category":"management","tagline":"Let your team launch pre-approved setups without asking an engineer.","description":"AWS Service Catalog is a menu of pre-approved cloud setups that your team can launch on their own. If you have contractors or new developers who need a test environment, you can publish a ready-made template (a saved recipe for building cloud resources) and let them pick it, instead of handing out broad access. It keeps everyone inside the rules you set. You pay per API call (a request to the service) after a free tier. Its Application Registry feature is in maintenance.","useCases":["Offer approved templates to teammates","Let teams launch pre-approved infrastructure","Share approved infrastructure templates"],"pricing":"Per API call after free tier.","status":"ga","url":"https://aws.amazon.com/servicecatalog/","source":"https://aws.amazon.com/products/"},{"id":"aws-service-quotas","provider":"aws","name":"Service Quotas","category":"management","tagline":"See your AWS usage limits and ask for higher ones in one place.","description":"Service Quotas is one page where you can see the limits AWS puts on your account and ask for them to be raised. If your app suddenly gets popular and you hit a cap on how many servers you can run, you can request more here instead of hunting through each service's settings. It is also a quick way to check your limits before a big launch so nothing surprises you. It is free to use.","useCases":["See current service limits in one place","Request increases to service limits","Track limits across all services"],"pricing":"Free.","status":"ga","url":"https://docs.aws.amazon.com/servicequotas/latest/userguide/intro.html","source":"https://aws.amazon.com/products/"},{"id":"aws-resource-explorer","provider":"aws","name":"AWS Resource Explorer","category":"management","tagline":"Search everything you have running in AWS, across all regions.","description":"AWS Resource Explorer is a search box for everything you have set up in AWS. If you have been experimenting with AI tools that create cloud resources and you can't remember what exists or where, you can search by name, tag (a label you attach to things) or type across all your accounts. It beats clicking through each service's page one by one. It is free.","useCases":["Search resources by name or tag","Find resources across all regions","Locate resources in every account"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/resourceexplorer/","source":"https://aws.amazon.com/products/"},{"id":"aws-resource-groups","provider":"aws","name":"AWS Resource Groups","category":"management","tagline":"Bundle related AWS resources by label so you can manage them together.","description":"AWS Resource Groups lets you collect resources into a group based on tags, which are simple labels like project=launch. If you run a side project and a client project in the same account, you can tag each one and then view or manage everything for a single project in one place. It is a good fit when you want label-based organization instead of hunting through each service separately. It is free.","useCases":["Group resources using tags","Act on tagged resources together","Organize resources by project tag"],"pricing":"Free.","status":"ga","url":"https://docs.aws.amazon.com/ARG/latest/userguide/resource-groups.html","source":"https://aws.amazon.com/products/"},{"id":"aws-resilience-hub","provider":"aws","name":"AWS Resilience Hub","category":"management","also":["observability"],"tagline":"Check whether your app can survive outages and recover quickly.","description":"AWS Resilience Hub checks how well your app would hold up if something in the cloud broke. If you promise customers that your product will be back within an hour of an outage, you set recovery targets (RTO is how long you can be down, RPO is how much recent data you can afford to lose) and it evaluates your setup against them. It then recommends improvements. Pricing is per application per month.","useCases":["Check if an app survives failures","Measure against recovery time targets","Get fixes to improve resilience"],"pricing":"Per application per month.","status":"ga","url":"https://aws.amazon.com/resilience-hub/","source":"https://aws.amazon.com/products/"},{"id":"aws-well-architected-tool","provider":"aws","name":"AWS Well-Architected Tool","category":"management","tagline":"A free guided checklist to review your setup against AWS best practices.","description":"The AWS Well-Architected Tool is a questionnaire that walks you through the Well-Architected Framework, AWS's set of best practices for building reliable, secure and affordable cloud apps. Before you launch a product or take on bigger customers, you answer questions about your setup and see where the gaps are. It is self-service, so you don't need to book a consultant. It is free.","useCases":["Review workloads against best practices","Walk through a guided self-review","Run self-service architecture reviews"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/well-architected-tool/","source":"https://aws.amazon.com/products/"},{"id":"aws-well-architected-agent","provider":"aws","name":"AWS Well-Architected Agent","category":"management","also":["agents"],"tagline":"An AI helper that reviews your AWS setup and suggests fixes.","description":"AWS Well-Architected Agent is an AI helper that reviews your cloud setup against the Well-Architected Framework, AWS's set of best practices, and proposes changes. If you built your app quickly with AI tools and aren't sure it is set up well, it can look it over and tell you what to improve. Pick it over the Well-Architected Tool when you would rather have the review done for you than answer questions yourself. Pricing is based on usage.","useCases":["Review workloads with an AI agent","Get proposed architecture changes","Check designs against best practices"],"pricing":"Usage-based.","status":"ga","url":"https://aws.amazon.com/architecture/well-architected/","source":"https://aws.amazon.com/products/"},{"id":"aws-devops-agent","provider":"aws","name":"AWS DevOps Agent","category":"observability","also":["agents"],"tagline":"An AI agent that investigates AWS problems and suggests fixes.","description":"AWS DevOps Agent is an AI assistant that investigates when something goes wrong in your AWS setup. If an alarm goes off at night because your app is slow or down, it looks through your telemetry (the data your app reports about its own health) and your code, works out the root cause, and suggests ways to fix it. That saves you from digging through screens you may not understand. It is a good pick if you don't have an engineer on call. Pricing is based on usage.","useCases":["Investigate alarms across telemetry and code","Find root causes of incidents","Suggest fixes to prevent repeat incidents"],"pricing":"Usage-based.","status":"ga","url":"https://aws.amazon.com/devops-agent/","source":"https://aws.amazon.com/products/"},{"id":"aws-managed-grafana","provider":"aws","name":"Amazon Managed Grafana","category":"observability","tagline":"Ready-made dashboards for watching how your app is doing.","description":"Amazon Managed Grafana gives you Grafana, a popular tool for building charts and dashboards, without having to set it up or run it yourself. If you want one screen that shows how fast your app is, how many errors it has and how busy it is, you can connect it to sources such as CloudWatch (AWS's built-in monitoring) and Prometheus (a metrics system). Pick it when you want good dashboards without maintaining the software. You pay per active user per month.","useCases":["Build operational dashboards without managing servers","Connect CloudWatch and Prometheus data","View metrics from several data sources"],"pricing":"Per active user per month.","status":"ga","url":"https://aws.amazon.com/grafana/","source":"https://aws.amazon.com/products/"},{"id":"aws-managed-prometheus","provider":"aws","name":"Amazon Managed Service for Prometheus","category":"observability","also":["containers"],"tagline":"Store and track your container metrics without running the system yourself.","description":"Amazon Managed Service for Prometheus runs Prometheus, a popular open-source system for tracking numbers about how your software is behaving, so you don't have to. If your app runs in containers (packaged apps) on Kubernetes (a tool for managing many containers), this is where you would keep metrics like request counts. It is built to scale as you grow, so you don't need to worry about it filling up. You pay per sample (a single data point) ingested and stored.","useCases":["Monitor Kubernetes cluster metrics","Store container metrics at scale","Use Prometheus without running servers"],"pricing":"Per sample ingested and stored.","status":"ga","url":"https://aws.amazon.com/prometheus/","source":"https://aws.amazon.com/products/"},{"id":"aws-proton","provider":"aws","name":"AWS Proton","category":"devtools","also":["containers"],"tagline":"Standard app templates for teams, now heading toward end of support.","description":"AWS Proton lets a platform team publish standard setups for containers and serverless apps (apps that run without you managing servers), which developers then deploy themselves. In a larger company, it keeps every team building in the same approved way. It is free, and you pay only for the resources you create. AWS has announced end of support in 2026, so avoid starting anything new on it.","useCases":["Publish standard stacks for developers","Let developers deploy approved templates","Maintain existing stacks until support ends"],"pricing":"Free. Pay for resources.","status":"deprecated","url":"https://aws.amazon.com/proton/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-auto-scaling","provider":"aws","name":"EC2 Auto Scaling","category":"vms","tagline":"Automatically add or remove servers as your traffic goes up and down.","description":"EC2 Auto Scaling adds or removes EC2 servers (virtual computers you rent from AWS) automatically, based on a schedule or on measurements like how busy they are. If your app gets a rush of visitors after a launch or a newsletter, it adds servers to keep things fast, then removes them when things calm down so you aren't paying for idle machines. Pick it when your traffic is unpredictable. The feature itself is free, and you pay for the servers it runs.","useCases":["Add servers when traffic rises","Remove servers when demand drops","Scale on schedules or metrics"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/ec2/autoscaling/","source":"https://aws.amazon.com/products/"},{"id":"aws-dlm","provider":"aws","name":"Amazon Data Lifecycle Manager","category":"block-storage","tagline":"Automatically back up your app's disks on a schedule and clean up old ones.","description":"Amazon Data Lifecycle Manager creates and deletes backups of your server disks for you, based on rules you set. These backups are called snapshots, which are saved copies of a disk at a moment in time. If your app runs on an AWS server and you want a fresh backup every night, with copies older than a month removed, you set that rule once and stop thinking about it. Pick it over running your own backup scripts when you want scheduling and cleanup handled inside AWS. The service itself is free, and you only pay for the snapshots it stores.","useCases":["Schedule automatic disk snapshots","Keep snapshots for set periods","Delete old snapshots automatically"],"pricing":"Free. Pay for snapshots.","status":"ga","url":"https://docs.aws.amazon.com/ebs/latest/userguide/snapshot-lifecycle.html","source":"https://aws.amazon.com/products/"},{"id":"aws-launch-wizard","provider":"aws","name":"AWS Launch Wizard","category":"devtools","tagline":"Step-by-step setup for big business software like SQL Server and SAP on AWS.","description":"AWS Launch Wizard walks you through setting up large business software on AWS, such as SQL Server (a widely used database) and SAP (business management software). If your company needs to run one of these and nobody on the team knows how to size the servers, the wizard asks about your needs and picks a sensible setup. It then builds everything following AWS best practices. Choose it over doing the setup by hand when the software is complex and mistakes are costly. It is free to use.","useCases":["Deploy SQL Server with guided setup","Size SAP systems for deployment","Follow best-practice enterprise app setups"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/launchwizard/","source":"https://aws.amazon.com/products/"},{"id":"aws-user-notifications","provider":"aws","name":"AWS User Notifications","category":"management","also":["messaging"],"tagline":"One place to get AWS alerts by email, chat or phone.","description":"AWS User Notifications gathers messages about what is happening in your AWS account and sends them where you will actually see them: email, team chat or your phone. If you want to know when something breaks or needs attention in your app, you can set it up once instead of checking several AWS screens. It is a good pick when you want a single, central place to manage who hears about what, rather than wiring up alerts service by service. It is free.","useCases":["Send AWS alerts to email","Push AWS alerts to mobile phones","Send operational alerts to your team"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/notifications/","source":"https://aws.amazon.com/products/"},{"id":"aws-q-chat-applications","provider":"aws","name":"Amazon Q Developer in chat applications","category":"devtools","also":["messaging"],"tagline":"Get AWS alerts in Slack or Teams, and run AWS commands right from chat.","description":"Amazon Q Developer in chat applications connects your AWS account to Slack or Microsoft Teams. Alerts show up in a channel, and you can run AWS commands without opening the AWS console. If your app has a problem at night, your team can see the alert in Slack, look into it and respond in the same conversation. Pick it when your team already lives in chat and you want everyone to see problems together. It was formerly called AWS Chatbot, and it is free.","useCases":["Triage AWS alerts inside Slack","Run AWS commands from Teams","Get AWS alerts in Teams"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/q/developer/operate/","source":"https://aws.amazon.com/products/"},{"id":"aws-sustainability","provider":"aws","name":"AWS Sustainability","category":"management","tagline":"See how much carbon your AWS usage produces.","description":"AWS Sustainability shows you the carbon emissions linked to your AWS usage. If investors, customers or your own team ask how green your product is, you can pull a report instead of guessing. It helps most when you need numbers for a sustainability report or want to see whether changes to your setup reduce your footprint. Because it reads from your existing AWS usage, there is nothing extra to build or install. It is free.","useCases":["Track emissions from your AWS usage","Review carbon footprint insights","Prepare emissions reports"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/sustainability/","source":"https://aws.amazon.com/products/"},{"id":"aws-telco-network-builder","provider":"aws","name":"AWS Telco Network Builder","category":"specialized","tagline":"Set up and run 5G phone network software on AWS.","description":"AWS Telco Network Builder automates the setup of 5G network functions on AWS. These are the software pieces that make up a mobile phone network. If you are a telecom company building the core of a 5G network, it deploys and manages those pieces for you instead of having engineers configure each one by hand. This is a specialized tool, so most app builders will never need it. You pay per network function instance, meaning for each running piece of network software.","useCases":["Deploy 5G network functions automatically","Manage telecom networks on AWS","Set up 5G core networks"],"pricing":"Per network function instance.","status":"ga","url":"https://aws.amazon.com/tnb/","source":"https://aws.amazon.com/products/"},{"id":"aws-service-management-connector","provider":"aws","name":"AWS Service Management Connector","category":"management","tagline":"Manage AWS from ServiceNow or Jira Service Management (being retired).","description":"AWS Service Management Connector links AWS to IT help-desk tools, specifically ServiceNow and Jira Service Management. If your company handles requests through one of those tools, it lets people manage AWS resources without leaving it. It was free to use. AWS has announced a sunset for this service, so you should not start new projects on it and should plan for an alternative if you already rely on it.","useCases":["Connect AWS to ServiceNow tickets","Manage AWS from Jira Service Management","Link AWS changes to ITSM workflows"],"pricing":"Free.","status":"deprecated","url":"https://aws.amazon.com/service-management-connector/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-marketplace","provider":"aws","name":"AWS Marketplace","category":"management","tagline":"Buy and sell software that runs on AWS, billed on your AWS invoice.","description":"AWS Marketplace is an online catalog of software, data and services from other companies that you can buy and have charged to your AWS bill. If you need a tool such as a security product or an AI model, you can find it, subscribe and start using it without setting up a new vendor account or payment method. It is handy when your company already has an AWS agreement and wants simple purchasing. Each seller sets their own prices, and the charges appear on your regular AWS invoice.","useCases":["Buy third-party software through AWS","Pay vendors on your AWS bill","Find third-party data and services"],"pricing":"Vendor-set prices on your AWS bill.","status":"ga","url":"https://aws.amazon.com/marketplace/","source":"https://aws.amazon.com/products/"},{"id":"aws-mgn-discovery","provider":"aws","name":"AWS Application Discovery Service","category":"migration","tagline":"Take stock of your own servers before moving them to AWS (closed to new users).","description":"AWS Application Discovery Service collects information about the servers in your own data center, including what they run and how they connect to each other. If a company wants to move its old systems to the cloud, this helps map out what exists before anything moves. That kind of map makes planning a migration much easier. It was free to use. The service is now closed to new customers, so it is only available to those who already use it.","useCases":["Inventory on-premises servers before migrating","Map server dependencies for migration planning","Collect server inventory data"],"pricing":"Free.","status":"closed","url":"https://aws.amazon.com/application-discovery/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-migration-hub","provider":"aws","name":"AWS Migration Hub","category":"migration","tagline":"Follow all your move-to-AWS projects on one dashboard (closed to new users).","description":"AWS Migration Hub is a single dashboard for tracking how a move to AWS is going. If your company is moving several systems and different teams use different tools, it shows progress for all of them in one place. That saves you from asking each team for updates. It was free to use. The service is now closed to new customers, so only existing users can keep using it.","useCases":["Track migration progress in one dashboard","See status of all migrations","Monitor migrations from a central view"],"pricing":"Free.","status":"closed","url":"https://aws.amazon.com/migration-hub/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-dms","provider":"aws","name":"AWS Database Migration Service (DMS)","category":"migration","also":["data-pipelines"],"tagline":"Move a database to AWS while your app keeps running.","description":"AWS Database Migration Service copies data from one database to another while your app stays online, so downtime is minimal. If your product runs on an older database and you want to move to a newer one on AWS, it copies everything over and keeps the two in sync until you switch. It can even move between different database types, and it has a serverless option that adjusts to the workload. Pick it over a manual export when you cannot afford long outages. You pay per hour of the copying server, or by DCU (a unit of serverless capacity).","useCases":["Move databases with minimal downtime","Copy ongoing database changes continuously","Replicate data between different database engines"],"pricing":"Per replication instance-hour or DCU.","status":"ga","url":"https://aws.amazon.com/dms/","source":"https://aws.amazon.com/products/"},{"id":"aws-sct","provider":"aws","name":"AWS Schema Conversion Tool","category":"migration","tagline":"Rewrite a database's structure and code to work with a different database.","description":"AWS Schema Conversion Tool translates the structure and code of one database so it works in another. A schema is the blueprint of how your data is organized. If you are moving from Oracle, a costly commercial database, to PostgreSQL, a popular free one, this tool converts the blueprint and much of the code so you do not rewrite it all by hand. Use it when you are switching between different kinds of databases rather than just copying data. It is free to use.","useCases":["Convert database schemas between engines","Convert Oracle databases to PostgreSQL","Convert database code for new engines"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/dms/schema-conversion-tool/","source":"https://aws.amazon.com/products/"},{"id":"aws-mainframe-modernization","provider":"aws","name":"AWS Mainframe Modernization","category":"migration","tagline":"Move old mainframe programs to the cloud and keep them running.","description":"AWS Mainframe Modernization helps companies move programs built for big, old mainframe computers onto AWS. If your company has decades of business software written in COBOL, an older programming language, this is a way to get it off the mainframe. You can refactor, which means rewriting the code for modern setups, or replatform, which means running it mostly as it is on new infrastructure. You pay per hour that an environment runs. The managed runtime experience is in maintenance, so keep that in mind before starting something new.","useCases":["Move COBOL workloads onto AWS","Refactor mainframe applications for the cloud","Run modernized COBOL apps on AWS"],"pricing":"Per environment-hour.","status":"ga","url":"https://aws.amazon.com/mainframe-modernization/","source":"https://aws.amazon.com/products/"},{"id":"aws-evs","provider":"aws","name":"Amazon Elastic VMware Service","category":"migration","also":["vms"],"tagline":"Run your existing VMware setup inside your own AWS network.","description":"Amazon Elastic VMware Service runs VMware Cloud Foundation, VMware's bundle of tools for managing virtual servers, on AWS. If your company runs its servers on VMware in its own data center and wants to leave, you can move those servers here without rebuilding every app. It runs on EC2 bare metal, meaning whole physical AWS machines reserved for you, inside your VPC, which is your private network on AWS. Pick it when you want to keep the VMware tools your team already knows. You pay per host-hour.","useCases":["Move VMware workloads to AWS","Keep existing VMware setups running","Run VMware inside your private network"],"pricing":"Per host-hour.","status":"ga","url":"https://aws.amazon.com/evs/","source":"https://aws.amazon.com/products/"},{"id":"aws-mediaconvert","provider":"aws","name":"AWS Elemental MediaConvert","category":"video","tagline":"Convert video files into formats that play on phones, TVs and browsers.","description":"AWS Elemental MediaConvert takes video files and converts them into the formats needed for streaming and broadcast. If your app lets people upload videos, they will arrive in all kinds of formats and sizes, and this turns them into versions that viewers can actually watch. It is built for video that is already recorded, often called video on demand, rather than live video. For live events, look at MediaLive instead. You pay per minute of video produced, and the rate depends on the tier you choose.","useCases":["Convert video files for streaming","Prepare videos for broadcast formats","Transcode on-demand video libraries"],"pricing":"Per minute of output by tier.","status":"ga","url":"https://aws.amazon.com/mediaconvert/","source":"https://aws.amazon.com/products/"},{"id":"aws-medialive","provider":"aws","name":"AWS Elemental MediaLive","category":"video","tagline":"Prepare live video for streaming to viewers as it happens.","description":"AWS Elemental MediaLive is a live video encoder, which means it compresses and converts a live camera feed so it can be sent out over the internet or to broadcast. If you are running a live event, a conference or a weekly show, this is the piece that takes your raw feed and turns it into streams for broadcast and streaming outputs. Pick it for live video, and use MediaConvert when your video is already a file. You pay per hour that a channel is running.","useCases":["Encode live video for broadcasts","Encode live events for streaming","Produce live broadcast outputs"],"pricing":"Per hour of channel.","status":"ga","url":"https://aws.amazon.com/medialive/","source":"https://aws.amazon.com/products/"},{"id":"aws-mediapackage","provider":"aws","name":"AWS Elemental MediaPackage","category":"video","tagline":"Package live and recorded video for streaming, with copy protection.","description":"AWS Elemental MediaPackage prepares your video for delivery to viewers and acts as its home base, called an origin. It packages live and on-demand video into HLS and DASH, two common streaming formats, and can add DRM, which is copy protection for paid or restricted content. If you stream a live event and want replays afterward, one service can handle both. It sits after an encoder like MediaLive in the chain. You pay per GB of video sent in and per GB sent out.","useCases":["Package live video into HLS and DASH","Protect streams with DRM","Serve as origin for streams"],"pricing":"Per GB ingested and egress.","status":"ga","url":"https://aws.amazon.com/mediapackage/","source":"https://aws.amazon.com/products/"},{"id":"aws-mediatailor","provider":"aws","name":"AWS Elemental MediaTailor","category":"video","tagline":"Put ads into video streams, personalized for each viewer.","description":"AWS Elemental MediaTailor personalizes video streams with ads. It does this on the server side, meaning ads are stitched into the stream before it reaches the viewer, so people see one continuous video instead of a separate ad player. If you run a streaming service or free channel that earns money from advertising, this lets you show different ads to different viewers. It can also assemble channels from existing videos. You pay per ad insertion and per minute.","useCases":["Insert ads into video streams","Personalize ads for each viewer","Assemble channels with server-side ads"],"pricing":"Per ad insertion and per minute.","status":"ga","url":"https://aws.amazon.com/mediatailor/","source":"https://aws.amazon.com/products/"},{"id":"aws-mediaconnect","provider":"aws","name":"AWS Elemental MediaConnect","category":"video","tagline":"Send live video reliably between your location and the cloud.","description":"AWS Elemental MediaConnect moves live video from one place to another, such as from a venue or studio up to the cloud. If you film an event on site and need the feed to reach your cloud tools without drops, this is the pipe that carries it. These incoming feeds are often called contribution feeds. It is for transporting video, not for preparing it for viewers, so you would pair it with an encoder like MediaLive. You pay per hour a flow is running, plus data transfer.","useCases":["Send live video between sites","Move live feeds into the cloud","Carry contribution feeds reliably to cloud"],"pricing":"Per flow hour and transfer.","status":"ga","url":"https://aws.amazon.com/mediaconnect/","source":"https://aws.amazon.com/products/"},{"id":"aws-elemental-inference","provider":"aws","name":"AWS Elemental Inference","category":"video","also":["ai-apps"],"tagline":"Use AI to clip, tag and reformat live and recorded video.","description":"AWS Elemental Inference runs machine learning, a kind of AI that learns from examples, on your video to do jobs like clipping and tagging. If you run a sports or creator platform, you could use it to pull out highlights automatically, or to crop wide video into vertical versions for phones. It works on both live video and files, so you can use it during an event or on your back catalog. Pricing is usage-based, so you pay for what you process.","useCases":["Create highlight clips from video","Crop video vertically for phones","Tag objects in live or file video"],"pricing":"Usage-based.","status":"ga","url":"https://aws.amazon.com/elemental-inference/","source":"https://aws.amazon.com/products/"},{"id":"aws-elemental-on-premises","provider":"aws","name":"AWS Elemental On-Premises","category":"video","tagline":"Video equipment and software you run in your own building.","description":"AWS Elemental On-Premises is a set of appliances, which are physical boxes, and software for processing video on site instead of in the cloud. If you operate a broadcast facility or a studio that needs to encode and handle video where the cameras are, this is the option for that. Pick it when your video has to be processed in your own building. If you are building a typical web or mobile app, the cloud video services in this catalog are usually a closer fit. Pricing is through contacting sales.","useCases":["Process video in on-site facilities","Equip broadcast facilities with encoders","Run video encoding appliances on site"],"pricing":"Contact sales.","status":"ga","url":"https://aws.amazon.com/elemental-appliances-software/","source":"https://aws.amazon.com/products/"},{"id":"aws-cdi-sdk","provider":"aws","name":"AWS Cloud Digital Interface SDK","category":"video","tagline":"Free toolkit for sending top-quality, uncompressed video between AWS servers.","description":"The AWS Cloud Digital Interface SDK is a free, open source toolkit, meaning code that developers add to their own software. It moves uncompressed video, which keeps full quality but makes very large files, between EC2 instances, the servers you rent from AWS. It is built for timing-critical work. If you are building tools for a broadcast studio that wants to produce shows in the cloud, this gives your software a way to pass pristine video around. It is a specialist tool and most apps will not need it.","useCases":["Transfer uncompressed video between instances","Build cloud broadcast production workflows","Use open source video transport code"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/media-services/resources/cdi/","source":"https://aws.amazon.com/products/"},{"id":"aws-ivs","provider":"aws","name":"Amazon Interactive Video Service (IVS)","category":"video","also":["realtime"],"tagline":"Add live streaming with a built-in player and chat to your app.","description":"Amazon Interactive Video Service, or IVS, is a managed way to add live streaming to your own app or site. If you want to run live shopping events or let creators stream to their fans, it gives you low-latency streaming, meaning very little delay between the broadcaster and the viewer, with players and chat built in. It also offers real-time stages, where several people can join and interact on video. Pick it when you want live video without building the plumbing yourself. You pay per hour of video in and out.","useCases":["Launch live shopping streams","Run creator live streams","Add real-time interactive video stages"],"pricing":"Per hour of video input and output.","status":"ga","url":"https://aws.amazon.com/ivs/","source":"https://aws.amazon.com/products/"},{"id":"aws-deadline-cloud","provider":"aws","name":"AWS Deadline Cloud","category":"video","also":["specialized"],"tagline":"Rent a render farm to produce animation and visual effects faster.","description":"AWS Deadline Cloud is a managed render farm. Rendering is the slow step of turning a 3D scene into finished images, and a render farm spreads that work across many computers at once. If you make animation or visual effects and your own machines cannot keep up before a deadline, you can scale up here and scale back down afterward. It works with common DCC tools, the software artists use to create 3D and effects work. You pay per vCPU-hour, which is an hour of processor power, plus any software licensing.","useCases":["Render animation and visual effects","Scale rendering across many machines","Run VFX pipelines with common tools"],"pricing":"Per vCPU-hour and licensing.","status":"ga","url":"https://aws.amazon.com/deadline-cloud/","source":"https://aws.amazon.com/products/"},{"id":"aws-ses","provider":"aws","name":"Amazon Simple Email Service (SES)","category":"email","tagline":"Send app emails like receipts and newsletters for $0.10 per 1,000.","description":"Amazon Simple Email Service (SES) lets your app send and receive email at large scale. If your app sends password resets, receipts or order confirmations, SES delivers them for $0.10 per 1,000 emails, and a free tier is available. It also handles marketing emails like newsletters. You get dedicated IPs (sending addresses used only by you), DKIM (a signature that proves emails really come from you) and tools that help your messages reach the inbox. Pick it when you expect to send a lot of email and want a low cost per message.","useCases":["Send transactional emails at scale","Send marketing email campaigns","Improve email deliverability with DKIM"],"pricing":"$0.10 per 1,000 emails. Free tier available.","status":"ga","url":"https://aws.amazon.com/ses/","source":"https://aws.amazon.com/products/"},{"id":"aws-pinpoint","provider":"aws","name":"Amazon Pinpoint","category":"messaging","tagline":"Run marketing campaigns by email, text and push. Being retired.","description":"Amazon Pinpoint helps you send marketing campaigns by email, text message (SMS) and push notifications (the alerts that pop up on a phone). If you want to welcome new users with a series of messages or send a promotion to a group of customers, campaigns and journeys (multi-step message flows) are what it was built for. Pricing is per message and per endpoint, meaning each device or address you contact. Support ends on October 30, 2026, and AWS End User Messaging is the replacement.","useCases":["Run email, SMS and push campaigns","Send messages across email, SMS and push","Message users across several channels"],"pricing":"Per message and endpoint.","status":"deprecated","url":"https://aws.amazon.com/pinpoint/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-eum-sms","provider":"aws","name":"AWS End User Messaging SMS","category":"messaging","tagline":"Send text messages and voice calls to your users around the world.","description":"AWS End User Messaging SMS sends text messages and voice calls from your app to people worldwide. If your app needs to send one-time login codes (often called OTP) or alerts like a delivery update, this is the tool for the job. It also helps you manage the phone numbers you send from. You pay per message, and the price depends on the country. Pick it when your app already runs on AWS and you want texting handled in the same place.","useCases":["Send one-time passcodes by text","Send app alerts by SMS","Send text and voice messages worldwide"],"pricing":"Per message by country.","status":"ga","url":"https://aws.amazon.com/end-user-messaging/","source":"https://aws.amazon.com/products/"},{"id":"aws-eum-push","provider":"aws","name":"AWS End User Messaging Push","category":"messaging","tagline":"Send push notifications to your iPhone and Android app users.","description":"AWS End User Messaging Push sends push notifications, the little alerts that pop up on a phone, to your iOS and Android apps. If you run a mobile app and want to tell people about a new message, an update or a reminder, this does the sending. It works through APNs and FCM, which are Apple's and Google's own notification delivery systems. You pay per message. Pick it when your app already runs on AWS and you want notifications managed alongside everything else.","useCases":["Send push alerts to iOS apps","Send push alerts to Android apps","Notify users of new activity"],"pricing":"Per message.","status":"ga","url":"https://aws.amazon.com/end-user-messaging/","source":"https://aws.amazon.com/products/"},{"id":"aws-eum-social","provider":"aws","name":"AWS End User Messaging Social","category":"messaging","tagline":"Message your customers on WhatsApp from your app.","description":"AWS End User Messaging Social lets your app send WhatsApp Business messages through AWS. If your customers already use WhatsApp, you can send them order updates or answer support questions there instead of relying only on email. Pricing is per message, plus WhatsApp's own fees, so check both when you estimate costs. Pick it when you want WhatsApp as a channel and prefer to manage it next to your other AWS services.","useCases":["Send WhatsApp messages to customers","Send order updates over WhatsApp","Message customers from your app"],"pricing":"Per message plus WhatsApp fees.","status":"ga","url":"https://aws.amazon.com/end-user-messaging/","source":"https://aws.amazon.com/products/"},{"id":"aws-connect","provider":"aws","name":"Amazon Connect Customer","category":"end-user","also":["ai-apps"],"tagline":"A cloud call center for phone, chat and tasks, with AI help for agents.","description":"Amazon Connect Customer is a contact center in the cloud, meaning a system for handling customer calls and chats. If you run a support team and want customers to reach you by phone or chat, it covers both, plus tasks for your team to work through. It includes AI agent assist, which helps your staff during conversations, and self-service IVR, the phone menus customers can use without talking to a person. Pricing is pay as you go, per minute of use and per message.","useCases":["Run a call center with chat","Handle calls with automated phone menus","Give support agents AI suggestions"],"pricing":"Per minute of use and per message.","status":"ga","url":"https://aws.amazon.com/connect/","source":"https://aws.amazon.com/products/"},{"id":"aws-connect-decisions","provider":"aws","name":"Amazon Connect Decisions","category":"end-user","also":["ai-apps"],"tagline":"AI help for planning how much demand to expect and how much to supply.","description":"Amazon Connect Decisions uses AI to help supply chain planners make decisions about demand (what customers will want) and supply (what you have available to meet it). If you sell physical products and spend your time guessing how much stock to line up, it supports that planning work. It is aimed at planning teams rather than app builders, so think of it as a business tool, not a building block for your app. Pricing is not listed, so you contact AWS sales.","useCases":["Forecast demand for products","Plan supply to match demand","Make stock and supply decisions"],"pricing":"Contact sales.","status":"ga","url":"https://aws.amazon.com/connect/","source":"https://aws.amazon.com/products/"},{"id":"aws-connect-health","provider":"aws","name":"Amazon Connect Health","category":"end-user","also":["ai-apps"],"tagline":"AI agents that help healthcare teams with patient contact and paperwork.","description":"Amazon Connect Health offers agentic AI, meaning AI that can carry out tasks on its own, for healthcare organizations. If you run a clinic or build a health product, these AI agents can help with patient contact and clinical administration, the routine paperwork around care. It is made for patient engagement in healthcare, so it fits better there than a general chat tool would. Pricing is not published, so you contact AWS sales.","useCases":["Handle patient calls and messages","Automate clinic administrative tasks","Support patients with AI agents"],"pricing":"Contact sales.","status":"ga","url":"https://aws.amazon.com/connect/","source":"https://aws.amazon.com/products/"},{"id":"aws-connect-talent","provider":"aws","name":"Amazon Connect Talent","category":"end-user","also":["ai-apps"],"tagline":"AI that helps recruiters screen and evaluate job candidates.","description":"Amazon Connect Talent uses AI to help recruiters screen and evaluate candidates. If your company gets a large pile of applications for each opening, it can help your hiring team work through them and judge who looks like a good fit. It is aimed at hiring teams rather than people building their own apps, so it works as a business tool you use, not a part you wire in. Pricing is not listed, so you contact AWS sales.","useCases":["Screen job candidates faster","Evaluate candidates with AI help","Support recruiters in hiring decisions"],"pricing":"Contact sales.","status":"ga","url":"https://aws.amazon.com/connect/","source":"https://aws.amazon.com/products/"},{"id":"aws-supply-chain","provider":"aws","name":"AWS Supply Chain","category":"end-user","tagline":"See your inventory and plan demand in one place, priced per user.","description":"AWS Supply Chain is a ready-made application that pulls your supply chain data together so you can see what is happening in one view. If you sell physical goods and your inventory information is scattered across different systems, it gives you a single picture of your stock and helps with demand planning. It uses machine learning, which is software that learns from patterns in data, to support this. Pricing is per user per month, so cost grows with the size of your team.","useCases":["See inventory across locations","Combine supply chain data in one place","Plan demand using smart forecasts"],"pricing":"Per user per month.","status":"ga","url":"https://aws.amazon.com/aws-supply-chain/","source":"https://aws.amazon.com/products/"},{"id":"aws-chime","provider":"aws","name":"Amazon Chime","category":"end-user","tagline":"A meetings and chat app from AWS, now retired.","description":"Amazon Chime was an AWS app for online meetings and chat, the kind of tool a team uses to hold video calls and message each other. If you want to add calling or video inside your own product, look at Amazon Chime SDK instead, which is a separate service. The Chime app was retired in February 2026, so it is no longer an option for new projects.","useCases":["Hold video meetings with teams","Chat with colleagues in one app","Schedule and join online meetings"],"pricing":"Retired.","status":"deprecated","url":"https://aws.amazon.com/chime/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-chime-sdk","provider":"aws","name":"Amazon Chime SDK","category":"realtime","also":["video"],"tagline":"Add video calls, voice calls and chat to your own app.","description":"Amazon Chime SDK gives developers building blocks (an SDK is a toolkit for adding a feature) to put audio, video and messaging inside your own app. If you are building a telehealth app where patients see a doctor on video, or you want calling built into your product, this covers it. You pay per attendee-minute, meaning for each person for each minute they are in a meeting. One date to know: the SIP Media Application, the part that handles regular phone calls, is moving to maintenance on October 29, 2026.","useCases":["Add audio and video calls to apps","Build voice calling inside an app","Power telehealth video visits"],"pricing":"Per attendee-minute.","status":"ga","url":"https://aws.amazon.com/chime/chime-sdk/","source":"https://aws.amazon.com/products/"},{"id":"aws-workmail","provider":"aws","name":"Amazon WorkMail","category":"email","also":["end-user"],"tagline":"Business email and calendar for your team, now being shut down.","description":"Amazon WorkMail is a hosted email and calendar service for your company, so you get business mailboxes without running your own mail software. If you are setting up a small team and want addresses like you@yourcompany.com, it gives each person a mailbox and calendar for a monthly fee per mailbox. It fits when you want your email to live inside AWS, the Amazon cloud, rather than with a separate email provider. AWS has announced the end of this service and it is in sunset, so choose a different email provider for anything new.","useCases":["Provide business email mailboxes","Share calendars across a team","Host company email addresses"],"pricing":"Per mailbox per month.","status":"deprecated","url":"https://aws.amazon.com/workmail/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-workspaces","provider":"aws","name":"Amazon WorkSpaces","category":"end-user","tagline":"A full Windows or Linux computer in the cloud, usable from any device.","description":"Amazon WorkSpaces gives people a complete desktop computer that lives in the cloud and shows up on whatever device they already own. If you bring on a contractor for a few months, you can set up a desktop for them instead of shipping a laptop, then remove it when they finish. It also suits remote teams who want the same desktop wherever they log in. You pay monthly or hourly per desktop, so short-term helpers do not need to cost a full month.","useCases":["Give remote staff a desktop","Set up desktops for contractors","Stream Windows desktops to any device"],"pricing":"Monthly or hourly per desktop.","status":"ga","url":"https://aws.amazon.com/workspaces/","source":"https://aws.amazon.com/products/"},{"id":"aws-workspaces-applications","provider":"aws","name":"Amazon WorkSpaces Applications","category":"end-user","tagline":"Let people run your Windows desktop apps in a browser, with no install.","description":"Amazon WorkSpaces Applications streams desktop programs to people's web browsers, so they can use the software without installing it. It was formerly called AppStream 2.0. If you have a Windows program and want to offer it as an online subscription without rewriting it, you can run it on AWS and let customers open it from a browser. Pick it over full virtual desktops when users only need one or two apps, not a whole computer. Pricing is per instance-hour, meaning per hour of the server running your app.","useCases":["Deliver Windows apps to users online","Stream software to customers in browsers","Run desktop apps without installing them"],"pricing":"Per instance-hour.","status":"ga","url":"https://aws.amazon.com/workspaces/applications/","source":"https://aws.amazon.com/products/"},{"id":"aws-workspaces-core","provider":"aws","name":"Amazon WorkSpaces Core","category":"end-user","tagline":"Cloud machines that run virtual desktop software from other companies.","description":"Amazon WorkSpaces Core provides the underlying AWS infrastructure for virtual desktop products from partners such as Citrix and Omnissa. VDI, short for virtual desktop infrastructure, means employees open a desktop that actually runs in a data center. If your company already uses one of those partner tools and wants to host it on AWS, this supplies the machines while you keep the software your team knows. Choose it over Amazon WorkSpaces when you want to stay with your partner's software. You pay per hour or per month.","useCases":["Host partner virtual desktop software","Run Citrix or Omnissa desktops","Manage desktop infrastructure for partners"],"pricing":"Per hour or month.","status":"ga","url":"https://aws.amazon.com/workspaces-family/workspaces/","source":"https://aws.amazon.com/products/"},{"id":"aws-workspaces-thin-client","provider":"aws","name":"Amazon WorkSpaces Thin Client","category":"end-user","tagline":"A small, low-cost device that connects people to their cloud desktops.","description":"Amazon WorkSpaces Thin Client is a small, low-cost device that connects to desktops running in the cloud. If you run a call center, you can give each agent one of these instead of a full laptop, since the real computing happens in the cloud. It is a good pick when people only need a way to reach their cloud desktop and nothing else on the machine in front of them. You pay per device, which makes costs easy to predict as your team grows.","useCases":["Equip call centers with desktop devices","Connect low-cost devices to cloud desktops","Use small devices for virtual desktops"],"pricing":"Per device.","status":"ga","url":"https://aws.amazon.com/workspaces-family/workspaces/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-workspaces-secure-browser","provider":"aws","name":"Amazon WorkSpaces Secure Browser","category":"zero-trust","also":["end-user"],"tagline":"Open internal sites and web apps in a locked-down browser in the cloud.","description":"Amazon WorkSpaces Secure Browser lets people reach company websites and web apps through a browser that runs in the cloud, so the pages never run directly on their own device. This is called remote browser isolation. If your contractors use their own laptops, they can open your internal tools this way without you handing out company computers, which is often called BYOD, short for bring your own device. It is priced per user per month. For new customers, it moves to maintenance mode from Oct 29, 2026.","useCases":["Let staff open internal sites safely","Let staff use personal devices securely","Isolate browsing to protect company data"],"pricing":"Per user per month.","status":"closed","url":"https://aws.amazon.com/workspaces-family/secure-browser/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-dcv","provider":"aws","name":"Amazon DCV","category":"end-user","tagline":"Smooth remote access to graphics-heavy computers, free on AWS.","description":"Amazon DCV is a remote display tool that lets you see and control a strong computer from far away, even when it is running 3D graphics. If your team does 3D work that needs a beefy machine, you can keep that machine in the cloud and connect from a regular laptop. It streams 3D desktops with high performance and costs nothing when you run it on EC2, the AWS service for renting servers. Pick it when picture quality and speed matter more than basic remote access.","useCases":["Stream 3D workstations to remote users","Run graphics apps from the cloud","Show high-performance desktops remotely"],"pricing":"Free on AWS.","status":"ga","url":"https://aws.amazon.com/hpc/dcv/","source":"https://aws.amazon.com/products/"},{"id":"aws-gamelift-servers","provider":"aws","name":"Amazon GameLift Servers","category":"specialized","also":["vms"],"tagline":"Host and scale the servers that run your multiplayer game.","description":"Amazon GameLift Servers runs the game servers that players connect to in an online multiplayer game, and adds more of them as players arrive. If you are building a game where friends play together, it also handles matchmaking, which means grouping players into the same match. Pick it over setting up servers yourself when you would rather not manage capacity by hand. You pay per instance-hour, which means per hour that each server is running.","useCases":["Host multiplayer game servers","Match players into online games","Scale game servers up and down"],"pricing":"Per instance-hour.","status":"ga","url":"https://aws.amazon.com/gamelift/servers/","source":"https://aws.amazon.com/products/"},{"id":"aws-gamelift-streams","provider":"aws","name":"Amazon GameLift Streams","category":"specialized","also":["video"],"tagline":"Stream games to a web browser with very little delay.","description":"Amazon GameLift Streams runs a game on a cloud computer with a graphics chip, called a GPU, and streams the picture to the player's browser. If you want people to try a demo of your game without downloading anything, you can let them play it straight from a web page. It also works for cloud gaming, where the game runs far away and the player just watches and controls it. Pick it when you want players to start on any device with no install. You pay per streaming hour.","useCases":["Stream games to web browsers","Let players try games instantly","Run game demos without downloads"],"pricing":"Per streaming hour.","status":"ga","url":"https://aws.amazon.com/gamelift/streams/","source":"https://aws.amazon.com/products/"},{"id":"aws-braket","provider":"aws","name":"Amazon Braket","category":"specialized","tagline":"Try quantum computing programs on real quantum machines or simulators.","description":"Amazon Braket lets you run quantum computing programs, a newer kind of computing that researchers use for certain very hard problems, without owning any quantum machine. If you are exploring quantum research, you can test an idea on a simulator first and then run it on hardware from IonQ, Rigetti or IQM. It puts several hardware makers in one service, so you can compare them without building separate setups. You pay per task plus per shot, where a shot is one run of your program.","useCases":["Run quantum algorithms on real hardware","Test quantum programs on simulators","Compare quantum hardware from several vendors"],"pricing":"Per task plus per shot.","status":"ga","url":"https://aws.amazon.com/braket/","source":"https://aws.amazon.com/products/","gem":"Run code on real quantum computers from three hardware makers."},{"id":"aws-ground-station","provider":"aws","name":"AWS Ground Station","category":"specialized","tagline":"Rent satellite dish time to talk to your satellites.","description":"AWS Ground Station lets you use satellite antennas on the ground without building your own. It contacts your satellite as it passes overhead and sends the data it collects into AWS. If you run a satellite that takes pictures of Earth, you can use it to receive those images, a step known as downlink. It is a good fit when building your own dish sites would be too costly. You pay per minute of antenna time, so you only pay while the antenna is in use.","useCases":["Download data from satellites","Rent satellite antenna time by minute","Send earth imaging data to cloud"],"pricing":"Per minute of antenna time.","status":"ga","url":"https://aws.amazon.com/ground-station/","source":"https://aws.amazon.com/products/","gem":"Rent a satellite antenna by the minute. Really."},{"id":"aws-managed-blockchain","provider":"aws","name":"Amazon Managed Blockchain","category":"specialized","tagline":"Run blockchain network nodes without managing the servers yourself.","description":"Amazon Managed Blockchain runs the computers, called nodes, that make up a blockchain, which is a shared record book that several organizations can trust. It supports Hyperledger Fabric and Ethereum, two popular blockchain systems. If a group of companies wants to share a tamper-resistant record, it sets up their nodes for them, billed per node-hour. Support for the service ends on Sept 29, 2027, so avoid it for new projects.","useCases":["Run shared blockchain networks","Operate Ethereum or Hyperledger nodes","Set up consortium blockchain projects"],"pricing":"Per node-hour.","status":"deprecated","url":"https://aws.amazon.com/managed-blockchain/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-iot-core","provider":"aws","name":"AWS IoT Core","category":"iot","also":["queues"],"tagline":"Let your connected devices send and receive messages through the cloud.","description":"AWS IoT Core is the messaging hub for physical devices like sensors, trackers and smart appliances. It uses MQTT, a lightweight messaging format built for small devices, plus a rules engine that sends each message where it needs to go. If your product has sensors that report temperature, you can collect those readings here and send commands back, like telling a device to switch off. Pick it when you want devices talking reliably to the rest of your AWS setup. You pay per message and per minute a device stays connected.","useCases":["Send sensor readings to the cloud","Send commands to connected devices","Route device messages with rules"],"pricing":"Per message and connection minute.","status":"ga","url":"https://aws.amazon.com/iot-core/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-device-defender","provider":"aws","name":"AWS IoT Device Defender","category":"iot","also":["threat-intel"],"tagline":"Check your connected devices for security gaps and odd behavior.","description":"AWS IoT Device Defender watches over a fleet of connected devices and checks that they are set up safely. If you ship thousands of smart gadgets, you can run audits to catch weak security settings and keep an eye out for devices acting strangely. Pick it when you already use AWS for your devices and want security checks designed for fleets rather than for single computers. Its Detect feature is in maintenance. Pricing is per device per month.","useCases":["Audit security across device fleets","Monitor unusual device behavior","Check devices for security gaps"],"pricing":"Per device per month.","status":"ga","url":"https://aws.amazon.com/iot-device-defender/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-device-management","provider":"aws","name":"AWS IoT Device Management","category":"iot","tagline":"Add, organize and update your connected devices from one place.","description":"AWS IoT Device Management helps you keep track of many connected devices after they leave the factory. If you sell smart gadgets and need to ship a bug fix, you can send a software update over the internet, called an OTA (over-the-air) update, to every device or just a chosen group. It also lets you search your fleet and run jobs, which are tasks sent to many devices at once. Choose it when updating devices by hand stops being realistic. Pricing is per device and per operation.","useCases":["Register and organize devices","Send software updates over the air","Run batch jobs across device fleets"],"pricing":"Per device and operation.","status":"ga","url":"https://aws.amazon.com/iot-device-management/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-expresslink","provider":"aws","name":"AWS IoT ExpressLink","category":"iot","tagline":"Ready-made modules that give your device a secure link to AWS.","description":"AWS IoT ExpressLink is a small hardware module you add to a device so it can connect to AWS without you writing all the networking code yourself. If you are building a gadget, like a smart sensor, and your team has little cloud experience, the module takes care of the secure connection for you. Pick it over building that connection from scratch on your own chip when you want to get to market faster. You buy the modules through partners.","useCases":["Connect embedded devices to AWS","Add a secure cloud link to hardware","Use partner modules for connectivity"],"pricing":"Via partner modules.","status":"ga","url":"https://aws.amazon.com/iot-expresslink/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-fleetwise","provider":"aws","name":"AWS IoT FleetWise","category":"iot","tagline":"Collect data from vehicles and send it to the cloud.","description":"AWS IoT FleetWise gathers data from vehicles, like speed or engine readings, and sends it to the cloud. If you run a delivery fleet and want to study how your vehicles perform, it collects those signals for analysis. It is designed for vehicles specifically, rather than for general sensors. Pricing is per signal, meaning you pay for each type of vehicle reading collected. This service is closed to new customers.","useCases":["Collect data from vehicle fleets","Transfer vehicle data to the cloud","Run analytics on fleet vehicles"],"pricing":"Per signal.","status":"closed","url":"https://aws.amazon.com/iot-fleetwise/","source":"https://docs.aws.amazon.com/general/latest/gr/maintenance_services.html"},{"id":"aws-iot-greengrass","provider":"aws","name":"AWS IoT Greengrass","category":"iot","also":["edge-compute"],"tagline":"Run your code right on the device, even when the internet is slow.","description":"AWS IoT Greengrass lets a device run software locally instead of sending everything to the cloud. It can run Lambda, AWS's small on-demand code snippets, and containers, which are packaged apps. If a camera in a factory needs to spot defects, it can run an AI model on the spot, known as edge inference, without waiting on a connection. Pick it when speed or spotty internet matters to your product. Pricing is per device per month.","useCases":["Run cloud code on local devices","Run AI models directly on devices","Run containers on local devices"],"pricing":"Per device per month.","status":"ga","url":"https://aws.amazon.com/greengrass/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-sitewise","provider":"aws","name":"AWS IoT SiteWise","category":"iot","tagline":"Collect and organize data from factory and industrial equipment.","description":"AWS IoT SiteWise gathers readings from machines in factories, plants and other industrial sites, then organizes them into models, which are digital descriptions of each machine. If you want factory metrics like how much each machine produced today, it brings them together in one place instead of leaving them in separate systems. Pick it when your data comes from industrial equipment rather than consumer gadgets. Its Monitor feature is in maintenance. You pay per message and for storage.","useCases":["Collect data from factory equipment","Model industrial equipment data","Track factory performance metrics"],"pricing":"Per message and storage.","status":"ga","url":"https://aws.amazon.com/iot-sitewise/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-twinmaker","provider":"aws","name":"AWS IoT TwinMaker","category":"iot","tagline":"Build a live digital copy of a building or factory.","description":"AWS IoT TwinMaker lets you build digital twins, which are virtual copies of real places like buildings and plants. If you manage a factory, you can create a virtual view of it that connects to live data from your sensors, such as which machine is running hot. Pick it when you want one place to see and ask questions about a whole site, rather than juggling separate dashboards. Pricing is per entity, meaning each item you model, and per query.","useCases":["Build digital twins of buildings","Build digital twins of plants","Model a building or plant virtually"],"pricing":"Per entity and query.","status":"ga","url":"https://aws.amazon.com/iot-twinmaker/","source":"https://aws.amazon.com/products/"},{"id":"aws-iot-wireless","provider":"aws","name":"AWS IoT Wireless","category":"iot","tagline":"Connect low-power wireless devices using LoRaWAN and Sidewalk.","description":"AWS IoT Wireless manages devices that talk over LoRaWAN and Sidewalk, two low-power wireless networks. These networks suit small devices that must run a long time on a battery and often sit far from Wi-Fi. If you place soil sensors across a farm, you can connect and manage them here. Pick it when Wi-Fi or cellular would be too costly or too short-range for your devices. You pay per message and per device.","useCases":["Connect low-power wireless devices","Manage LoRaWAN device fleets","Manage Sidewalk-connected devices"],"pricing":"Per message and device.","status":"ga","url":"https://aws.amazon.com/iot-core/lorawan/","source":"https://aws.amazon.com/products/"},{"id":"aws-freertos","provider":"aws","name":"FreeRTOS","category":"iot","tagline":"Free, tiny operating system for the small chips inside devices.","description":"FreeRTOS is a free, open source operating system for microcontrollers, the tiny low-cost chips inside things like thermostats and wearables. An RTOS, or real-time operating system, makes sure tasks happen exactly on time. If you are writing firmware, the software that runs on that chip, it gives you a base to build on. AWS adds libraries that make connecting to its services easier. Pick it when your device is too small to run a full system like Linux. It costs nothing.","useCases":["Program microcontroller firmware","Run real-time tasks on microcontrollers","Connect microcontrollers to AWS"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/freertos/","source":"https://aws.amazon.com/products/"},{"id":"aws-partner-central","provider":"aws","name":"AWS Partner Central","category":"management","tagline":"The portal for managing your AWS partner membership and co-selling.","description":"AWS Partner Central is a website for companies that belong to the AWS Partner Network, Amazon's program for businesses that build on or resell AWS. If your startup sells software that runs on AWS, you can manage your membership and partner programs there. It also supports co-selling, where you and AWS work together to win customers. Pick it when you are an AWS partner and want one home for those tasks. It is for managing your partnership, not for running apps. It is free.","useCases":["Manage your AWS partner membership","Manage partner programs","Co-sell with AWS"],"pricing":"Free.","status":"ga","url":"https://aws.amazon.com/partners/partner-central/","source":"https://aws.amazon.com/products/"},{"id":"azure-microsoft-foundry","provider":"azure","name":"Microsoft Foundry","category":"ai-platform","tagline":"One place to build, launch and oversee AI apps and agents.","description":"Microsoft Foundry, formerly Azure AI Foundry and Azure AI Studio, is a single workspace for building AI apps and agents, which are AI helpers that carry out tasks. It brings together a catalog of models, tools and controls for governing how AI is used. If you want a chatbot that answers from your company documents, a pattern called RAG, you can build it here, test the answers and deploy it. Pick it when you want everything in one place rather than stitching together separate tools. You pay for the underlying models and services you use.","useCases":["Build AI apps and agents","Choose and compare AI models","Evaluate and deploy AI agents"],"pricing":"Pay for the underlying models and services used.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-foundry-models","provider":"azure","name":"Foundry Models","category":"ai-models","tagline":"Browse, compare and deploy thousands of AI models from one place.","description":"Foundry Models is a catalog of AI models from Microsoft and partners such as OpenAI, Meta and Mistral, where you can pick one and deploy it. If you are building a chatbot and are not sure which model gives the best answers for the price, you can compare several here before you commit. You can pay per token (a small chunk of text) with serverless models, where you only pay for what you use, or per hour with managed compute, which is a server set aside for you. Choose it when you want many options in one place. It was formerly the Azure AI model catalog.","useCases":["Compare models from many providers","Call models through serverless APIs","Fine-tune models on your data"],"pricing":"Per token for serverless models; per hour for managed compute.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/models/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-openai-in-foundry-models","provider":"azure","name":"Azure OpenAI in Foundry Models","category":"ai-models","tagline":"OpenAI's GPT and other models, hosted on Azure with business controls.","description":"Azure OpenAI in Foundry Models runs OpenAI's models inside Azure, Microsoft's cloud. If you are building a chat assistant or a tool that summarizes long documents, you get the same kind of models behind ChatGPT, plus private networking, content filtering and regional data residency (keeping your data in a region you choose). You can pay as you go per token, or reserve steady capacity with provisioned throughput. Pick it over using OpenAI directly when your company or customers care about where data lives and who can reach it. It was formerly Azure OpenAI Service.","useCases":["Build chat assistants","Summarize long documents","Generate code with AI"],"pricing":"Per token, or provisioned throughput units.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/models/openai/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-phi-open-models","provider":"azure","name":"Phi open models","category":"ai-models","tagline":"Small, low-cost AI models you can run in the cloud or on a device.","description":"Phi is Microsoft's family of small language models, which are quicker and cheaper to run than the giant ones. If your app handles a huge number of simple requests, like tagging messages or writing short replies, or needs an assistant that runs on a phone or other device, Phi is a good fit. The models are open-weight, meaning the model files are free to download and use. You can also call them through Foundry, paying per token, or get them from Hugging Face, a popular site for sharing models. Choose Phi over a larger model when speed and cost matter more than top-end ability.","useCases":["Run small AI models on devices","Serve high-volume AI requests cheaply","Power assistants at the edge"],"pricing":"Per token on Foundry; free open weights.","status":"ga","url":"https://azure.microsoft.com/en-us/products/phi/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-foundry-agent-service","provider":"azure","name":"Foundry Agent Service","category":"agents","tagline":"Run AI agents that use tools and your files, without building the plumbing.","description":"Foundry Agent Service hosts AI agents for you. An agent is an AI that can take steps and use tools, not just chat. If you want a customer support agent that searches your help files, runs a quick calculation and kicks off a workflow, this gives you the pieces: conversation threads, tool calling, file search, a code interpreter and connections to Logic Apps, Azure's workflow automation tool. Pick it when you would rather not build and run all that machinery yourself. You pay for model tokens and tool usage. It was formerly Azure AI Agent Service.","useCases":["Build customer support agents","Automate multi-step workflows","Analyze data with AI agents"],"pricing":"Pay for model tokens and tool usage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/agent-service/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-foundry-iq","provider":"azure","name":"Foundry IQ","category":"ai-search","tagline":"Let your AI agents look things up in your company's data, with permissions.","description":"Foundry IQ is a knowledge layer that lets AI agents find answers in your organization's information. If you are building an internal assistant that should answer from your wikis, documents and other sources, it gives every agent one shared way to look things up, and it respects who is allowed to see what. That cuts the work of building a RAG pipeline (where an AI searches your documents before answering) separately for each agent. It is built on Azure AI Search, and you are billed for usage through that service and the models.","useCases":["Give agents access to company data","Search data across enterprise sources","Skip building custom retrieval pipelines"],"pricing":"Usage-based; billed through Azure AI Search and models.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/iq/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-foundry-control-plane","provider":"azure","name":"Foundry Control Plane","category":"ai-gateway","tagline":"One dashboard to track, secure and set rules for all your AI agents.","description":"Foundry Control Plane is a central place to see and manage the AI agents your organization runs. If several teams have built agents with different tools and frameworks, you can list them all, set policies, manage their identities, check safety and monitor the whole fleet from one spot. It is aimed at companies running many agents, so a solo project with a single agent may not need it. It comes with Foundry, with usage-based add-ons available for extra features.","useCases":["Track every AI agent in one place","Set policies for AI agents","Monitor agents built on different frameworks"],"pricing":"Included with Foundry; usage-based add-ons.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/control-plane","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-observability-in-foundry-control-plane","provider":"azure","name":"Observability in Foundry Control Plane","category":"observability","tagline":"See what your AI app did, how well it answered and what it cost.","description":"Observability in Foundry Control Plane shows you what is happening inside your AI apps and agents. If your agent gives a strange answer, you can open its trace, a step-by-step record of the run, to see what went wrong. It also collects quality and safety checks, called evaluations, plus cost numbers, so you can spot problems and spending early. It works with Azure Monitor and Application Insights, Azure's standard monitoring tools. Choose it when you want checks designed for generative AI apps. You pay based on usage through Azure Monitor.","useCases":["Trace what AI agent runs do","Score AI output quality continuously","Track AI app costs"],"pricing":"Usage-based through Azure Monitor.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/observability","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-content-safety-in-foundry-control-plane","provider":"azure","name":"Content Safety in Foundry Control Plane","category":"ai-gateway","tagline":"Spot harmful text and images, and block tricks aimed at your AI.","description":"Content Safety checks text and images for harmful material, whether people wrote it or an AI generated it. If you run a public chatbot, you can screen what users send, filter what the AI replies and block jailbreaks, which are prompts that try to trick an AI into ignoring its rules. It also offers groundedness detection, which checks whether an AI answer is actually backed by the source material you gave it. Pricing is per 1,000 text records or per image, with a free tier. It was formerly Azure AI Content Safety.","useCases":["Moderate user chat messages","Detect prompt attacks on AI apps","Filter harmful generated images"],"pricing":"Per 1,000 text records or per image; free tier available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-services/ai-content-safety/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-foundry-tools","provider":"azure","name":"Foundry Tools","category":"ai-apps","tagline":"Ready-made AI for reading images, transcribing speech and translating text.","description":"Foundry Tools is the umbrella name for Microsoft's prebuilt AI services, formerly Azure AI Services and, before that, Cognitive Services. Each one is an API (a way for your app to ask for a job to be done) that handles a single task without any model training. If your app lets people snap photos of receipts, you can add OCR, which reads the text in an image, with one call. Pick these over a general AI model when you want a dedicated tool for one job. You pay per transaction, and most have a free tier.","useCases":["Add text recognition from images","Transcribe audio without training models","Translate text with prebuilt APIs"],"pricing":"Per transaction; most have a free tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-language-in-foundry-tools","provider":"azure","name":"Azure Language in Foundry Tools","category":"ai-apps","tagline":"Find sentiment, names, summaries and personal info in text.","description":"Azure Language in Foundry Tools analyzes written text and pulls out useful details. If you collect customer reviews or call notes, you can score how positive or negative they are, summarize them and redact PII, meaning personal details like names and phone numbers. It also offers custom classification, question answering and conversational language understanding, which works out what a person wants from their message. Choose it when you want a dedicated text analysis tool with predictable pricing, charged per 1,000 text records, with a free tier. It was formerly Azure AI Language and Text Analytics.","useCases":["Detect sentiment in customer feedback","Redact personal data from text","Summarize long call transcripts"],"pricing":"Per 1,000 text records; free tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/language","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-translator-in-foundry-tools","provider":"azure","name":"Azure Translator in Foundry Tools","category":"ai-apps","tagline":"Translate text and whole documents across 100+ languages.","description":"Azure Translator in Foundry Tools translates text and files between more than 100 languages. If customers write to your support team in many languages, you can translate their tickets as they arrive, or run a batch of documents through at once to localize them. Custom terminology and custom models let you keep product names and specialist words consistent. Pick it when you want translation as its own dedicated service. Pricing is per million characters, with a free tier. It was formerly Azure AI Translator.","useCases":["Translate support tickets quickly","Localize documents into many languages","Use custom terminology for translation"],"pricing":"Per million characters; free tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/translator","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-speech-in-foundry-tools","provider":"azure","name":"Azure Speech in Foundry Tools","category":"ai-apps","tagline":"Turn speech into text and text into natural-sounding voices.","description":"Azure Speech in Foundry Tools handles talking and listening for your app. If you want to transcribe customer calls, build a voice assistant or turn written chapters into an audiobook, it offers real-time and batch transcription, speech translation and neural voices, which are natural-sounding AI voices. Custom neural voice lets you build a voice of your own, but access is gated, so you have to be approved first. Pricing is per audio hour or per million characters, with a free tier. It was formerly Azure AI Speech.","useCases":["Transcribe customer calls","Add voice output to apps","Produce audiobook narration with voices"],"pricing":"Per audio hour or per million characters; free tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/speech/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-vision-in-foundry-tools","provider":"azure","name":"Azure Vision in Foundry Tools","category":"ai-apps","tagline":"Let your app read text in photos, tag images and write captions.","description":"Azure Vision in Foundry Tools is a set of ready-made services that look at pictures and tell you what is in them. If your app lets people photograph receipts, it can read the text off the image, a skill known as OCR (optical character recognition). It can also tag pictures, write captions for screen reader users, and help analyze video. Pick it when you want image understanding without training your own model. It used to be called Azure AI Vision and Computer Vision.","useCases":["Read text from receipts","Tag and caption pictures","Caption images for accessibility"],"pricing":"Per 1,000 transactions; free tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/vision","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-document-intelligence-in-foundry-tools","provider":"azure","name":"Azure Document Intelligence in Foundry Tools","category":"ai-apps","tagline":"Pull text, tables and fields out of invoices, forms and IDs.","description":"Azure Document Intelligence in Foundry Tools reads documents and forms and hands back the information inside as organized data. If you run a bookkeeping app, you can upload an invoice and get back the vendor, date and total without anyone typing them in. It comes with ready-made models for invoices, receipts and IDs, and you can train a custom one for your own contracts or forms. Pick it over general image reading when your files follow a known layout and you need specific fields. It was formerly Azure Form Recognizer.","useCases":["Extract fields from invoices","Verify IDs from documents","Pull data from contracts"],"pricing":"Per page; free tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/document-intelligence","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-content-understanding-in-foundry-tools","provider":"azure","name":"Azure Content Understanding in Foundry Tools","category":"ai-apps","tagline":"Turn documents, calls, video and images into clean, organized data.","description":"Azure Content Understanding in Foundry Tools uses generative AI to read mixed content, like documents, audio, video and images, and return the details you ask for in a set structure. If you run a support team, you could feed in call recordings and get back fields such as the customer's issue and how the call ended. You define the fields, and every result follows that shape. It is aimed at feeding AI agents and RAG (retrieval-augmented generation, where an AI looks up your data before answering). Pick it when your content comes in many formats.","useCases":["Summarize video content","Analyze call center recordings","Extract fields from forms"],"pricing":"Usage-based by content processed.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-foundry/tools/content-understanding","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-ai-search","provider":"azure","name":"Azure AI Search","category":"ai-search","also":["vector"],"tagline":"Add smart search to your app, including search that understands meaning.","description":"Azure AI Search lets people search through your content, whether that is documents, products or a whole website. If you want a chatbot that answers questions from your company files, those files get indexed (organized for fast lookup) here so the AI can find the right passages first. This pattern is called RAG, short for retrieval-augmented generation. It supports vector search, which matches by meaning instead of exact words, and hybrid search, which combines both. It plugs into Azure OpenAI and Foundry agents, and has a free tier with limits. It was formerly Azure Cognitive Search.","useCases":["Search company data for AI apps","Add site search to websites","Power product search"],"pricing":"Hourly by tier; free tier with limits.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-services/ai-search","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-ai-bot-service","provider":"azure","name":"Azure AI Bot Service","category":"ai-apps","tagline":"Build chatbots and put them in Teams, on your website and more.","description":"Azure AI Bot Service is the hosting and connection layer for chatbots. If you want a support bot that answers customers on your website and also inside Microsoft Teams, it handles the links to each of those places, called channels. You build the bot itself with the Bot Framework or the Microsoft 365 Agents SDK (developer toolkits), and Bot Service delivers it. Standard channels are free, and premium channels are billed by message. Pick it when you want one bot reachable in many places. It was formerly Azure Bot Service.","useCases":["Build Teams bots","Connect bots to Teams and web","Add chat to websites"],"pricing":"Free standard channels; premium channels billed by message.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-services/ai-bot-service/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-healthcare-agent-service","provider":"azure","name":"Healthcare agent service","category":"ai-apps","tagline":"Build AI health assistants with healthcare safeguards built in.","description":"Healthcare agent service is a platform for building virtual health assistants that chat with patients. If you run a clinic app, you could offer symptom triage, appointment scheduling or answers about benefits. It includes healthcare safeguards, can give generative AI answers based on your own sources, and works with Copilot Studio, Microsoft's tool for building assistants. Pick it over a general chatbot builder when compliance in a health setting matters. Billing is by message volume. It was formerly Azure Health Bot.","useCases":["Triage patient symptoms in a health assistant","Schedule patient appointments through chat","Answer benefits questions from your documents"],"pricing":"Billed by message volume; see the billing rates page.","status":"ga","url":"https://learn.microsoft.com/en-us/azure/health-bot/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-ai-video-indexer","provider":"azure","name":"Azure AI Video Indexer","category":"ai-apps","tagline":"Make your videos and audio searchable by spoken words, faces and text.","description":"Azure AI Video Indexer watches and listens to your stored videos and pulls out what is in them: spoken words, text on screen, faces and who is speaking. If you run a training platform with hundreds of recorded lessons, it can make every one searchable and create captions and transcripts. It can also help flag content that needs moderation. Pick it when you want these insights from finished recordings without building your own tools. It runs in the cloud or at the edge through Azure Arc (a way to run Azure services on your own equipment). You pay per minute indexed, with a free trial.","useCases":["Make video libraries searchable","Generate captions and transcripts automatically","Moderate uploaded video content"],"pricing":"Pay per minute of content indexed; free trial account.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-video-indexer/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-ai-custom-vision","provider":"azure","name":"Azure AI Custom Vision","category":"ai-apps","tagline":"Train your own model to sort images or spot objects in photos.","description":"Azure AI Custom Vision lets you teach a model to recognize things specific to your business by showing it example photos. If you run a factory, you could train it to spot defective parts on a production line, or to recognize your products on a shelf. It handles image classification (labeling a whole picture) and object detection (finding items within a picture). This service is being retired in favor of Azure Vision and Foundry. New resource creation is already restricted, and retirement is scheduled for 2028.","useCases":["Spot defects in product photos","Recognize products in images","Train a model on your own pictures"],"pricing":"Per transaction plus training hours.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/ai-services/ai-custom-vision/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-ai-immersive-reader","provider":"azure","name":"Azure AI Immersive Reader","category":"ai-apps","tagline":"Reading aids like read aloud and translation you can embed in your app.","description":"Azure AI Immersive Reader is a reading helper you can build into your own app. If you make an education app, students can have text read aloud, translated or made easier to follow, which helps learners and readers with dyslexia. It is a quick way to add accessibility features without creating them yourself. You pay per transaction. Microsoft has announced the retirement of the standalone service, so check its plans before you build on it.","useCases":["Read text aloud in learning apps","Translate text inside reading tools","Add reading support for accessibility"],"pricing":"Per transaction.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ai-services/ai-immersive-reader/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-machine-learning","provider":"azure","name":"Azure Machine Learning","category":"ai-platform","tagline":"A workspace to train, deploy and manage your own machine learning models.","description":"Azure Machine Learning is a workspace where data teams build, train and run their own machine learning models. If you have unique data and want a model that predicts which customers are about to leave, you can write code in notebooks, use AutoML (which tries different model approaches for you) and train across many GPUs (chips that speed up AI training). You can then deploy the model as a managed endpoint, a web address your app can call. Pick it when ready-made AI services do not fit your problem. The service adds no charge; you pay for the computing you use.","useCases":["Train custom models on your data","Deploy models as live endpoints","Run repeatable machine learning pipelines"],"pricing":"No extra charge; pay for compute.","status":"ga","url":"https://azure.microsoft.com/en-us/products/machine-learning/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-databricks","provider":"azure","name":"Azure Databricks","category":"warehouse","also":["ai-platform"],"tagline":"Crunch huge amounts of data and run analytics, built into Azure.","description":"Azure Databricks is a joint Microsoft and Databricks service for working with very large amounts of data. If your app has collected millions of customer events, you can clean and reshape them (known as ETL), query them with SQL and train machine learning models, all in one place. It is built on a lakehouse, which blends cheap file storage with fast warehouse-style queries, and on Apache Spark, an engine for big data. It connects with Entra ID (Microsoft's sign-in system) and Power BI. You pay for DBU usage (Databricks' usage unit) plus the servers.","useCases":["Process large datasets with Spark","Run analytics on a lakehouse","Train machine learning models on data"],"pricing":"DBU usage plus VM costs.","status":"ga","url":"https://azure.microsoft.com/en-us/products/databricks/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-open-datasets","provider":"azure","name":"Azure Open Datasets","category":"data-pipelines","tagline":"Free public datasets like weather and census data for AI and analysis.","description":"Azure Open Datasets is a library of ready-to-use public data, such as weather, census and mobility (how people move around) data. If you are building a model to predict delivery times, you can add weather data to your own without hunting for it and cleaning it up. It is also handy for research and prototyping. The datasets plug into Azure Machine Learning and Synapse, Azure's analytics service. The datasets are free to use; you pay only for the computing and storage that process them.","useCases":["Add public weather and census data","Enrich training data with public sets","Test ideas with ready-made public data"],"pricing":"Datasets are free to use; you pay for the compute and storage that processes them.","status":"ga","url":"https://azure.microsoft.com/en-us/products/open-datasets/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-data-science-virtual-machines","provider":"azure","name":"Data Science Virtual Machines","category":"vms","tagline":"Cloud computers that come with data science and AI tools already installed.","description":"Data Science Virtual Machines are ready-made virtual computers, in Windows or Linux, that already have popular data science, machine learning and deep learning tools installed. If you want to try training a model for your product, you can start one up and get to work instead of spending a day installing software. They also suit classrooms and workshops, where everyone needs the same setup. There is no extra charge for the images themselves. You only pay for the computing power and storage you use.","useCases":["Launch ready-made data science workstations","Run training experiments on virtual machines","Set up teaching labs quickly"],"pricing":"No extra charge; pay for the VM compute and storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machines/data-science-virtual-machines/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-security-copilot","provider":"azure","name":"Microsoft Security Copilot","category":"ai-apps","tagline":"An AI assistant that helps security teams investigate threats and incidents.","description":"Microsoft Security Copilot is an AI assistant for the people who defend a company's systems from attacks. If your team gets an alert about something suspicious, it can help investigate the incident, summarize the threat and draft a report. It works across Microsoft's security tools, including Defender, Sentinel, Entra and Purview. Pick it if you already use those tools and want AI help inside them. You pay per security compute unit hour.","useCases":["Investigate security incidents with AI","Write queries to hunt threats","Draft security incident reports"],"pricing":"Per security compute unit hour.","status":"ga","url":"https://www.microsoft.com/en-us/security/business/ai-machine-learning/microsoft-security-copilot","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-copilot","provider":"azure","name":"Azure Copilot","category":"ai-apps","tagline":"An AI helper inside Azure that answers questions and fixes problems.","description":"Azure Copilot is an AI helper built into the Azure portal, the website where you manage your cloud setup. If a server is misbehaving, you can ask it what is wrong and it will help you troubleshoot. It can also write queries and scripts for you, including KQL, the query language Azure uses to search logs, and give advice on cutting costs. It was formerly called Microsoft Copilot in Azure. It is included for Azure users.","useCases":["Troubleshoot virtual machine problems","Write queries and scripts","Get tips to cut costs"],"pricing":"Included for Azure users.","status":"ga","url":"https://azure.microsoft.com/en-us/products/copilot","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-sre-agent","provider":"azure","name":"Azure SRE Agent","category":"agents","tagline":"An AI agent that investigates outages and can fix them, with your approval.","description":"Azure SRE Agent is an AI agent that watches over your Azure setup and acts when something goes wrong. When an alert fires in the middle of the night, it can dig through logs, work out the likely cause and suggest or run a fix. Approval controls let you decide what it may do on its own. Choose it if you have a live product but no dedicated operations team to handle incidents. Pricing is usage-based, in agent units.","useCases":["Investigate alerts and find causes","Propose or run fixes on resources","Handle routine operations tasks"],"pricing":"Usage-based agent units.","status":"ga","url":"https://azure.microsoft.com/en-us/products/sre-agent","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-infrastructure-resiliency-manager","provider":"azure","name":"Azure Infrastructure Resiliency Manager","category":"management","tagline":"Set goals for staying online and test whether your setup can meet them.","description":"Azure Infrastructure Resiliency Manager helps you decide how well your app should survive when something breaks, then checks whether it does. You set targets across your workloads and test them against failures and disaster recovery, which is the plan for getting back online after a major outage. If customers depend on your product being up, it gives you evidence instead of hope. It is currently in preview, meaning it is not yet fully released. For pricing, contact sales or see the product page.","useCases":["Set targets for infrastructure resiliency","Test workloads against failures","Validate disaster recovery plans"],"pricing":"Contact sales or see product page.","status":"preview","url":"https://azure.microsoft.com/en-us/products/infrastructure-resiliency-manager","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-analysis-services","provider":"azure","name":"Azure Analysis Services","category":"warehouse","tagline":"A cloud engine that turns raw business data into models reports can use.","description":"Azure Analysis Services hosts semantic models, which are shared definitions of your business numbers, like revenue or active customers, so every report calculates them the same way. If several dashboards keep showing different totals, a model like this gives them one source of truth. It uses tabular models compatible with SQL Server Analysis Services, so it suits teams moving an existing setup to the cloud. For new work, Microsoft steers people toward Power BI and Fabric semantic models. Billing is hourly by tier.","useCases":["Build semantic data models","Serve data models to reports","Power business intelligence reports"],"pricing":"Hourly by tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/analysis-services/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-data-explorer","provider":"azure","name":"Azure Data Explorer","category":"warehouse","tagline":"Search huge piles of logs and sensor data in seconds.","description":"Azure Data Explorer is a database built for logs, sensor readings and anything else that arrives with a timestamp. If your app produces millions of events a day and you want to ask questions about them, it can scan billions of records in seconds using KQL, a query language made for this. It uses the same technology that powers Azure Monitor and Fabric Real-Time Intelligence. Pick it over a general-purpose database when speed on huge event data matters. You pay per hour for the servers it runs on, plus storage.","useCases":["Query billions of logs in seconds","Analyze device sensor telemetry","Analyze time series data"],"pricing":"Per cluster hour plus storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/data-explorer/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-data-factory","provider":"azure","name":"Azure Data Factory","category":"data-pipelines","tagline":"Move data between your tools and reshape it automatically.","description":"Azure Data Factory moves data from one place to another and reshapes it along the way, a job often called ETL (extract, transform, load). If your sales numbers sit in one system and your analytics live in another, you can build a pipeline that copies them over automatically. You can build visually or with code, and it has more than 90 connectors to popular systems. Microsoft's newer path for this work is Fabric Data Factory. You pay per activity run and per hour of data movement.","useCases":["Copy data between systems","Build ETL data pipelines","Move existing SSIS packages to Azure"],"pricing":"Per activity run and data movement hour.","status":"ga","url":"https://azure.microsoft.com/en-us/products/data-factory/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-data-lake-storage","provider":"azure","name":"Azure Data Lake Storage","category":"object-storage","also":["warehouse"],"tagline":"Cloud file storage with real folders, built for big analytics projects.","description":"Azure Data Lake Storage is a place to keep huge amounts of data files, organized in folders like the ones on your computer. It is built on Blob Storage, Azure's basic file storage, and adds real folders, fine-grained permissions and compatibility with Hadoop, a common big-data toolset. If you are collecting lots of raw data for analysis, it keeps things tidy and quick to work with. It is the default store for the analytics tools Synapse, Databricks and Fabric. You pay per GB stored plus operations.","useCases":["Store data lake files in folders","Hold data for Databricks and Synapse","Power big data analytics jobs"],"pricing":"Per GB stored plus operations.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/data-lake-storage/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-data-share","provider":"azure","name":"Azure Data Share","category":"data-pipelines","tagline":"Share data with other companies and see who has received it.","description":"Azure Data Share lets you send data to other organizations while keeping control over it. If you want to give a partner a regular copy of your numbers, you choose what is shared, who gets it and the terms of use, and you can see every sharing relationship in one place. It supports snapshots, which are copies sent on a schedule, and in-place sharing, where the other side reads your data without making a copy. You pay as you go, based on snapshot runs and data movement.","useCases":["Share data with partner organizations","Send scheduled data snapshots","Track who receives shared data"],"pricing":"Pay as you go based on snapshot execution and data movement.","status":"ga","url":"https://azure.microsoft.com/en-us/products/data-share/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-stream-analytics","provider":"azure","name":"Azure Stream Analytics","category":"data-pipelines","also":["queues"],"tagline":"Spot patterns in live data as it arrives, using simple SQL-style queries.","description":"Azure Stream Analytics looks at data the moment it arrives, such as sensor readings or app events, and reacts right away. If you want a live dashboard or an alert when something odd happens, you write a query in a SQL-like language and it does the watching. It reads from Event Hubs, IoT Hub and Kafka, which are services that carry streams of events, and writes results to databases, storage or Power BI. It can run in the cloud or on edge devices. You pay per streaming unit hour, with no upfront cost.","useCases":["Build live dashboards from streams","Send alerts on sensor readings","Flag fraud signals as they happen"],"pricing":"Pay per streaming unit hour; no upfront cost.","status":"ga","url":"https://azure.microsoft.com/en-us/products/stream-analytics/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-synapse-analytics","provider":"azure","name":"Azure Synapse Analytics","category":"warehouse","tagline":"One service for data warehousing, big data processing and data pipelines.","description":"Azure Synapse Analytics brings several analytics tools together in one place: dedicated SQL pools (a data warehouse, which is a database built for reporting), serverless SQL that runs queries without you managing servers, Spark for crunching big data, and pipelines. If you need to combine data from many sources and answer business questions at scale, it covers both always-on and pay-as-you-go styles. Microsoft positions Fabric as the successor for new projects, so look there first if you are starting fresh. Pricing is per DWU hour (a unit of warehouse capacity) or per TB processed.","useCases":["Run an enterprise data warehouse","Run Spark analytics on large data","Query data with serverless SQL"],"pricing":"Per DWU hour or per TB processed.","status":"ga","url":"https://azure.microsoft.com/en-us/products/synapse-analytics/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-event-hubs","provider":"azure","name":"Event Hubs","category":"queues","also":["data-pipelines"],"tagline":"Catch huge streams of events from apps and devices as they happen.","description":"Event Hubs is a service that takes in a constant flood of data, like clicks, sensor readings or app activity, and holds it so other tools can process it. If your app tracks every click on your website and you want to analyze that activity live, Event Hubs is where those events land first. It speaks the Apache Kafka protocol (a widely used standard for streaming data), so a team already using Kafka can switch over without rewriting everything. It can also save a copy of your data to storage automatically. Pick it when you want streaming without running the servers yourself.","useCases":["Collect telemetry from devices and apps","Process clickstream data from websites","Save streams to storage automatically"],"pricing":"Pay as you go per throughput or processing unit; free trial.","status":"ga","url":"https://azure.microsoft.com/en-us/products/event-hubs/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-hdinsight","provider":"azure","name":"HDInsight","category":"warehouse","tagline":"Run big-data tools like Hadoop, Spark and Kafka without managing the setup.","description":"HDInsight lets you run popular open-source big data tools, such as Hadoop, Spark (for crunching huge datasets) and Kafka (for streaming data), as ready-made clusters on Azure. If your company already has a pile of data jobs built on these tools and wants to move them to the cloud, you can bring them over mostly as they are. It is mainly suited to moving an existing big data setup. If you are starting fresh and just want answers from your data, a simpler analytics service will likely fit better.","useCases":["Move Hadoop workloads to Azure","Run Spark batch jobs","Run managed Kafka streaming clusters"],"pricing":"Pay per cluster node hour; no upfront cost.","status":"ga","url":"https://azure.microsoft.com/en-us/products/hdinsight/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-power-bi-embedded","provider":"azure","name":"Power BI Embedded","category":"warehouse","tagline":"Put interactive Power BI charts and reports inside your own app.","description":"Power BI Embedded lets you show Microsoft's interactive reports and dashboards inside your own product, so you do not have to build charts from scratch. If you run a SaaS app and want customers to see their own usage stats, or you have a customer portal that needs reports, you can drop them right in. You pay for the computing capacity you use, which is billed separately from per-user Power BI licenses. Pick it when you want polished analytics in your app quickly, rather than building every chart yourself.","useCases":["Add reports inside your own app","Show dashboards to customers","Give customers analytics without building charts"],"pricing":"Pay for the capacity you use, with no upfront cost.","status":"ga","url":"https://azure.microsoft.com/en-us/products/power-bi-embedded/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-fabric","provider":"azure","name":"Microsoft Fabric","category":"warehouse","also":["data-pipelines","ai-platform"],"tagline":"One place for data storage, analysis, live dashboards and reports.","description":"Microsoft Fabric is an all-in-one analytics platform that brings data engineering, data warehousing, live data and Power BI reports under a single roof and a single bill. If you want to pull data from several sources, clean it up and show it in dashboards, you can do all of it in one place instead of stitching tools together. It is built on OneLake, a shared storage layer, so every tool reads the same data. It is also the path forward from Azure Synapse, an older Microsoft analytics service. A free trial is available.","useCases":["Build a lakehouse for analytics","Create real-time dashboards","Run enterprise business intelligence reports"],"pricing":"Capacity-based (Fabric SKUs); free trial.","status":"ga","url":"https://www.microsoft.com/en-us/microsoft-fabric","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-purview","provider":"azure","name":"Microsoft Purview","category":"management","tagline":"Find, label and protect your data, and prove you follow the rules.","description":"Microsoft Purview helps you keep track of what data your company has, where it lives and who can see it. If you handle customer information and need to show you are following privacy rules, Purview can catalog your data, tag sensitive items and trace where data came from. It also includes tools to stop sensitive information from leaking out. It combines features that used to be separate Microsoft products, Azure Purview and Azure Information Protection. Pick it when governance and compliance matter as much as storing the data.","useCases":["Catalog data and track lineage","Classify sensitive data automatically","Meet data compliance requirements"],"pricing":"Usage and license-based.","status":"ga","url":"https://www.microsoft.com/en-us/security/business/risk-management/microsoft-purview-data-governance","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-planetary-computer-pro","provider":"azure","name":"Microsoft Planetary Computer Pro","category":"data-pipelines","tagline":"Organize and analyze your own satellite and map data at large scale.","description":"Microsoft Planetary Computer Pro is a platform for storing, cataloging and analyzing geospatial data, meaning anything tied to places on a map, like satellite images. If your business needs to compare locations, such as picking the best site for a new store or tracking land changes over time, you can load your own imagery and run analysis on it. It supports STAC, a common standard for organizing location data, and is built to work with AI-based analysis. Pick it when your map data is private and you need it organized for business use.","useCases":["Catalog satellite imagery collections","Analyze geospatial data at scale","Support site selection decisions"],"pricing":"Usage-based; see the pricing page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/planetary-computer-pro","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-energy-data-services","provider":"azure","name":"Microsoft Energy Data Services","category":"specialized","tagline":"A managed home for energy industry data like seismic and well records.","description":"Microsoft Energy Data Services, now branded Azure Data Manager for Energy, is a cloud platform built for oil, gas and other energy companies. It runs OSDU, an open industry standard for organizing energy data. If your company has years of seismic surveys and well records scattered across old systems, this gives you one modern place to store and analyze them. It is a specialist tool, so it makes sense mainly for energy businesses. Billing is hourly, based on the size of the instance you choose.","useCases":["Store and manage seismic survey data","Modernize well and exploration data","Speed up energy data analytics"],"pricing":"Billed hourly by instance tier; see the pricing page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/data-manager-for-energy","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-chaos-studio","provider":"azure","name":"Azure Chaos Studio","category":"devtools","tagline":"Break things on purpose to find weak spots before real outages do.","description":"Azure Chaos Studio lets you deliberately cause controlled problems in your Azure setup, like shutting down a server, slowing the network or overloading the processor, to see how your app copes. If you are about to launch and want to know what happens when a server suddenly disappears, you can test it safely first. It is useful for rehearsal days where your team practices handling failures. Pick it when you would rather find weak points on your own schedule than during a real outage. You pay per minute of actions run.","useCases":["Test how systems handle failures","Run outage drills before launch","Check fixes before a release"],"pricing":"Pay as you go per action minute; Azure free account covers trials.","status":"ga","url":"https://azure.microsoft.com/en-us/products/chaos-studio/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-app-service","provider":"azure","name":"App Service","category":"hosting","tagline":"Host your website, API or app backend without managing servers.","description":"App Service is a managed place to run your web app, API or mobile backend, so you do not have to set up or maintain servers yourself. If you built a web app in Node.js, Python, Java, .NET or PHP, or packed it in a container, you can upload it and have it online with your own domain. It handles growing traffic with autoscaling and includes deployment slots, which let you test a new version before it goes live. Pick it when you want a standard, dependable home for a dynamic web app. A free tier is available.","useCases":["Host web apps and APIs","Scale apps up with demand","Add custom domains to sites"],"pricing":"Hourly by plan size; a free tier is available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/app-service/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-static-web-apps","provider":"azure","name":"Static Web Apps","category":"hosting","also":["serverless"],"tagline":"Host your website and its small backend, deployed straight from GitHub.","description":"Static Web Apps hosts the front end of your site, the pages people see, and spreads it around the world so it loads quickly. If you built a marketing site or a single-page app and want it live every time you save changes in GitHub, this does that automatically. It also includes sign-in and lets you add backend features using Azure Functions, small pieces of code that run on demand. It can build preview versions of your site for review. Pick it over App Service when your site is mostly pages with light backend needs. There is a free plan.","useCases":["Host static websites worldwide","Publish frontends with API backends","Spin up preview versions for changes"],"pricing":"Free plan; Standard per app per month.","status":"ga","url":"https://azure.microsoft.com/en-us/products/app-service/static/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-quantum","provider":"azure","name":"Azure Quantum","category":"specialized","tagline":"Try out real quantum computers and simulators from several companies.","description":"Azure Quantum is a cloud workspace where you can run programs on quantum computers, a new kind of machine that solves certain problems very differently, without owning any hardware. If you are a researcher or curious team testing a new algorithm, you can write it in Q# or Qiskit (languages for quantum programs) and try it on machines from IonQ, Quantinuum, Rigetti and others. It also estimates how many resources a program would need. Pick it for research and experiments rather than everyday app building. Some free credits are available.","useCases":["Run quantum programs on real hardware","Test quantum algorithms on simulators","Estimate resources for quantum programs"],"pricing":"Per provider usage; some free credits.","status":"ga","url":"https://azure.microsoft.com/en-us/solutions/quantum-computing/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-spot-virtual-machines","provider":"azure","name":"Azure Spot Virtual Machines","category":"vms","tagline":"Rent spare Azure servers at a big discount for jobs that can be paused.","description":"Azure Spot Virtual Machines let you rent servers from Azure's unused capacity at a low price. The catch is that Azure can take them back when it needs the space, so your work may be interrupted. If you run big batch jobs, like processing a pile of videos, or test runners that build your code, a stop and restart is no problem and the savings add up. The price changes over time but is often far below normal pay-as-you-go rates. Pick it for work that can handle interruptions, not for your live website.","useCases":["Run batch jobs on spare capacity","Run build machines at lower cost","Process jobs that tolerate interruptions"],"pricing":"Variable Spot price, often far below pay-as-you-go.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machines/spot/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-spring-apps","provider":"azure","name":"Azure Spring Apps","category":"containers","tagline":"Hosting for Java Spring Boot apps, closed to new customers.","description":"Azure Spring Apps hosts apps written with Spring Boot, a popular framework for building Java apps, and looks after the servers for you. If your company already runs a Spring Boot backend here, it keeps running and scaling without you managing machines. For a new project, Microsoft recommends Container Apps (runs packaged apps without server work) or AKS (Azure's managed Kubernetes, a tool for running many app containers). It closed to new customers on March 17, 2025, and retires on March 31, 2028.","useCases":["Keep existing Spring Boot apps running","Support Spring apps already deployed","Plan a move to Container Apps"],"pricing":"Existing customers only.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/spring-apps/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-vmware-solution","provider":"azure","name":"Azure VMware Solution","category":"vms","tagline":"Move your existing VMware setup to Azure and keep your tools.","description":"Azure VMware Solution lets a company move its existing VMware setup, a common way of running many virtual computers, into Azure's data centers without rebuilding it. If you inherited a VMware-based system, or a business customer is closing its own data center, you can move everything over and keep using the same VMware tools. It runs on dedicated physical machines rather than shared ones, and VMware verifies and supports it. Pick it over a regular server when the goal is moving existing systems quickly, not building something new.","useCases":["Move VMware servers to Azure","Set up VMware disaster recovery","Add capacity during busy periods"],"pricing":"Billed per host node; reservations available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-vmware/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-batch","provider":"azure","name":"Batch","category":"workflows","tagline":"Split big computing jobs across many servers, then shut them down.","description":"Batch runs big piles of computing work for you, splitting a job across many servers at once. If you run a video app that needs to convert hundreds of uploaded clips, you describe the tasks and Batch sets up the servers, schedules the work and scales them as needed. Compared with managing your own servers, you skip the scheduling headaches. Batch itself costs nothing. You pay only for the servers and storage it uses.","useCases":["Render images and video at scale","Run large financial simulations","Transcode media files in parallel"],"pricing":"No charge for Batch itself; pay for the VMs and storage used.","status":"ga","url":"https://azure.microsoft.com/en-us/products/batch/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-cloud-services","provider":"azure","name":"Cloud Services","category":"vms","tagline":"Older Azure hosting for web apps and background workers, retiring.","description":"Cloud Services is an older Azure way to host web apps and background workers without managing the operating system. If you inherited an older Azure app built with web and worker roles, this is likely what it runs on. For new work, Microsoft points to App Service (simple web app hosting), Container Apps or AKS (ways to run packaged apps). The original classic version retired on August 31, 2024. The extended support version is deprecated and retires on March 31, 2027.","useCases":["Keep legacy web apps running","Keep older background workers running","Plan a move to App Service"],"pricing":"Pay as you go for compute; existing deployments only.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/cloud-services/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-linux-virtual-machines","provider":"azure","name":"Linux Virtual Machines","category":"vms","tagline":"Rent a Linux computer in the cloud: Ubuntu, Red Hat, SUSE and more.","description":"Linux Virtual Machines gives you a computer in the cloud running Linux, which you control fully. If you need a web server, a database or a test box for your app, you can start from ready-made images of Ubuntu, Red Hat, SUSE, Oracle and more, with optional vendor support. Choose it over simpler hosting when you need to install your own software or tweak the system itself. You pay per second based on the size you pick, with lower prices if you reserve ahead.","useCases":["Run web servers on Linux","Host databases on Linux machines","Set up dev and test machines"],"pricing":"Pay per second by VM size; reservations available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machines/linux/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-sql-server-on-azure-virtual-machines","provider":"azure","name":"SQL Server on Azure Virtual Machines","category":"sql","tagline":"Run the full SQL Server database on a cloud server you control.","description":"SQL Server on Azure Virtual Machines runs the full Microsoft SQL Server database on a cloud server that you control. If your business already has an app built on SQL Server and you want it in the cloud without rewriting anything, this is the closest match to what you have today. Unlike Azure's fully managed database options, you get access to the operating system underneath, which some older apps need. Automated backup and patching are available as options. You pay for the server plus the SQL Server license.","useCases":["Move existing SQL Server databases to cloud","Run SQL Server with full system access","Host database apps needing server control"],"pricing":"VM compute plus SQL Server license, pay as you go or Hybrid Benefit.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machines/sql-server/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-virtual-machine-scale-sets","provider":"azure","name":"Virtual Machine Scale Sets","category":"vms","tagline":"Groups of identical servers that grow and shrink with your traffic.","description":"Virtual Machine Scale Sets manages a group of identical cloud servers and adds or removes them automatically as demand changes. If your app gets a traffic spike after a launch or a newsletter, it adds servers to keep things fast, then trims them back later. It handles up to thousands of Linux and Windows servers and can spread them across availability zones, which are separate data centers in one region. Pick it over a single server when your load varies. There is no extra charge beyond the servers themselves.","useCases":["Scale identical servers automatically with demand","Run large pools of worker machines","Run large groups of computing machines"],"pricing":"No extra charge; pay for the underlying VMs.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machine-scale-sets/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-virtual-machines","provider":"azure","name":"Virtual Machines","category":"vms","tagline":"Rent Windows or Linux computers in the cloud, in many sizes.","description":"Virtual Machines are rentable computers in Azure's data centers, running Windows or Linux, in many sizes. If your product needs a custom server, a machine with a graphics chip (GPU) for AI work, or an existing app moved over unchanged, this is the basic building block. You get full control of the operating system, the flexibility that simpler hosting trades away. You pay per second by size, and savings plans and reservations (lower prices for committing ahead) are available.","useCases":["Move existing servers into the cloud","Build custom servers with full control","Run workloads that need graphics chips"],"pricing":"Pay per second by size; savings plans and reservations available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machines","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-windows-server","provider":"azure","name":"Windows Server","category":"vms","tagline":"Run Windows Server apps in the cloud, reusing licenses you already own.","description":"Windows Server on Azure lets you run Windows-based business software in the cloud instead of on your own hardware. If you rely on an older Windows app, a file server or Active Directory (the system that manages staff logins), you can move it over, including Windows Server 2025. Azure Hybrid Benefit lets you apply Windows licenses you already own to lower the bill. Pick it over a Linux server when your software only runs on Windows.","useCases":["Run older Windows business apps","Host Active Directory login servers","Run shared Windows file servers"],"pricing":"VM compute plus Windows license; Azure Hybrid Benefit applies.","status":"ga","url":"https://azure.microsoft.com/en-us/products/windows-server/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-dedicated-host","provider":"azure","name":"Azure Dedicated Host","category":"vms","tagline":"A whole physical server just for you, to run your cloud machines on.","description":"Azure Dedicated Host gives you a whole physical server that only your organization uses, and you run your Azure virtual machines on it. If a customer or regulator requires that your data not share hardware with anyone else, this is how you meet that. It also lets you choose when maintenance happens and reuse software licenses you already own, known as bring your own license. Pick it over regular virtual machines only when you need that isolation or control. You are billed for each hour the host runs.","useCases":["Keep workloads on dedicated physical servers","Meet strict compliance isolation rules","Reuse existing server software licenses"],"pricing":"Billed per dedicated host hour.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-machines/dedicated-host/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-vm-image-builder","provider":"azure","name":"Azure VM Image Builder","category":"devtools","tagline":"Build a ready-to-go server template once and reuse it everywhere.","description":"Azure VM Image Builder creates ready-made server templates, called images, so every new server starts with your software and settings already in place. If your team launches many servers and is tired of setting each one up by hand, you describe the setup once and it builds the image for you. It is built on HashiCorp Packer, a popular image-building tool, and saves results to Azure Compute Gallery, Azure's storage place for images. The service is free. You pay only for temporary resources used during the build.","useCases":["Create standard machine images automatically","Patch and rebuild server images regularly","Share images through a central gallery"],"pricing":"Free service; pay for temporary build resources.","status":"ga","url":"https://azure.microsoft.com/en-us/products/image-builder/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-virtual-desktop","provider":"azure","name":"Azure Virtual Desktop","category":"end-user","tagline":"Windows desktops and apps in the cloud, open from any device.","description":"Azure Virtual Desktop gives people a Windows desktop and apps that run in the cloud and that they open from any device. If you have contractors or a remote team who need secure access to company tools, you can hand them a desktop without shipping laptops. It supports multi-session Windows, meaning several people share one machine to save money, and FSLogix, which keeps each person's settings between sessions. Access can be included with eligible Windows or Microsoft 365 licenses, so you mostly pay for the machines and storage.","useCases":["Give remote staff Windows desktops","Set up desktops for contractors","Stream apps securely to users"],"pricing":"No extra service fee with eligible licenses; pay for VMs and storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-desktop/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-kubernetes-service-aks","provider":"azure","name":"Azure Kubernetes Service (AKS)","category":"containers","tagline":"Run your app's containers on Azure while Azure manages the hard parts.","description":"Azure Kubernetes Service (AKS) runs your containers, which are packaged apps that behave the same anywhere, on Kubernetes, a popular system for running many containers together. Azure looks after the control plane, the part that coordinates everything. If your product is split into several small services, or you need to serve a machine learning model, AKS gives you a place to run and scale them. Pick it over simpler options when you want the full flexibility of Kubernetes. AKS Automatic goes further and also manages the servers and add-ons for you. The control plane has a free tier, and you pay for the servers you use.","useCases":["Run microservices across many containers","Serve machine learning models","Build internal platforms for teams"],"pricing":"Free control plane tier; paid uptime SLA tier; pay for nodes.","status":"ga","url":"https://azure.microsoft.com/en-us/products/kubernetes-service/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-functions","provider":"azure","name":"Azure Functions","category":"serverless","also":["workflows"],"tagline":"Run small pieces of code only when something happens.","description":"Azure Functions runs small pieces of code in response to events, such as a web request, a message landing in a queue, or a timer, with no servers for you to manage. If your app needs to send a welcome email after signup or run a cleanup job every night, you write just that bit of code. It beats running a full server when work arrives in bursts, because you pay per execution and get a monthly free grant. Durable Functions handle multi-step workflows that remember where they left off, and the Flex Consumption plan adds fast scaling.","useCases":["Run code when events happen","Process jobs from message queues","Schedule recurring background tasks"],"pricing":"Per execution and GB-second; monthly free grant.","status":"ga","url":"https://azure.microsoft.com/en-us/products/functions/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-container-instances","provider":"azure","name":"Azure Container Instances","category":"containers","tagline":"Run a container on Azure without setting up servers or clusters.","description":"Azure Container Instances runs a single container, a packaged app that behaves the same anywhere, with no cluster to set up or look after. If you need a one-off job, like processing a batch of files, or a build agent for your code pipeline, you start it, let it finish, and stop paying. Billing is per second for the CPU and memory you use. Pick it over Kubernetes when you want the simplest path to running a container. Each container also gets hypervisor isolation, which keeps it separate from other customers' workloads.","useCases":["Launch a container without servers","Run short burst jobs","Run simple container apps"],"pricing":"Pay per second for vCPU and memory used.","status":"ga","url":"https://azure.microsoft.com/en-us/products/container-instances/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-container-apps","provider":"azure","name":"Azure Container Apps","category":"containers","also":["serverless"],"tagline":"Run your app's containers and scale to zero, with no servers to manage.","description":"Azure Container Apps is a fully managed place to run containers, which are packaged apps that behave the same anywhere, without handling servers yourself. If your product is made of several small services, or reacts to events like new messages, you deploy them and they grow with demand. When nothing is happening they can scale to zero, and you pay per second of usage with a monthly free grant. It is built on Kubernetes, but you do not have to work with it directly. Pick it over AKS when you want less upkeep. It also supports jobs, GPUs and AI agent workloads.","useCases":["Run microservices without managing clusters","Run event-driven apps in containers","Scale containers down to zero"],"pricing":"Pay per second for usage; monthly free grant included.","status":"ga","url":"https://azure.microsoft.com/en-us/products/container-apps/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-nutanix-cloud-clusters","provider":"azure","name":"Nutanix Cloud Clusters","category":"vms","tagline":"Run your existing Nutanix setup on Azure's physical servers.","description":"Nutanix Cloud Clusters (NC2) on Azure extends Nutanix, a system companies use to manage their own servers and storage, into Azure. If your business is moving out of its own data center, you can bring the same setup across instead of rebuilding everything. It also fits disaster recovery, meaning a backup site if your main one fails, and extra capacity during busy periods. This is aimed at teams that already use Nutanix. You pay for the Azure bare metal nodes, which are physical servers, plus Nutanix licensing.","useCases":["Move Nutanix workloads into Azure","Set up cloud disaster recovery","Extend existing Nutanix systems to cloud"],"pricing":"Azure bare metal node cost plus Nutanix licensing.","status":"ga","url":"https://azure.microsoft.com/en-us/products/nutanix","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-linux","provider":"azure","name":"Azure Linux","category":"containers","tagline":"Microsoft's own lightweight operating system for running containers.","description":"Azure Linux is an operating system, the base software a server runs on, built and supported by Microsoft for running containers on Azure. If you run your app on AKS, Azure Linux can be the system on each server behind the scenes, and you rarely have to think about it. It has a small footprint, meaning fewer pieces installed, which suits secure container hosts. It also works on Azure virtual servers. Choose it when you want a Microsoft-supported default instead of picking and maintaining your own. There is no extra charge when you use it with AKS.","useCases":["Use as the OS for Kubernetes nodes","Run secure container host machines","Run lightweight Linux for containers"],"pricing":"No extra charge; included with AKS.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-linux/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-app-configuration","provider":"azure","name":"App Configuration","category":"devtools","tagline":"One place to keep your app's settings and on/off switches.","description":"App Configuration is a central store for your app's settings and feature flags, which are switches that turn a feature on or off without releasing new code. If you run several services that share the same settings, you change a value once instead of editing each one. If you want to show a new feature to only some users first, a flag lets you do that. It supports labels, snapshots and references to Key Vault, Azure's store for secrets. There are Free and Standard tiers, and Standard is billed per day and per request.","useCases":["Turn features on and off","Share settings across services","Store app settings in one place"],"pricing":"Free and Standard tiers; Standard billed per day and per request.","status":"ga","url":"https://azure.microsoft.com/en-us/products/app-configuration/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-red-hat-openshift","provider":"azure","name":"Azure Red Hat OpenShift","category":"containers","tagline":"Managed OpenShift, run jointly by Microsoft and Red Hat.","description":"Azure Red Hat OpenShift gives you managed OpenShift clusters, Red Hat's version of Kubernetes, which is a system for running many containers together. If your team already builds on OpenShift, or your company standardizes on it, you can run those same workloads on Azure without operating the clusters yourself. Microsoft and Red Hat provide joint support, so you are not stuck between two vendors. It includes OpenShift Virtualization and Confidential Containers for extra protection of sensitive workloads. You pay for worker nodes, which are the servers running your apps, plus an OpenShift fee.","useCases":["Run container apps on managed clusters","Get joint support from Microsoft and Red Hat","Run virtual machines alongside containers"],"pricing":"Pay for worker nodes plus an OpenShift fee.","status":"ga","url":"https://azure.microsoft.com/en-us/products/openshift/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-container-registry","provider":"azure","name":"Azure Container Registry","category":"containers","tagline":"A private storage shelf for your app's container images.","description":"Azure Container Registry is a private place to store your container images, the packaged versions of your app that get deployed, along with other build files such as Helm charts. If your team builds a new version of your app every day, this is where those versions live until a server picks them up. It works with any container deployment, not only Azure's. You also get geo-replication, which keeps copies in several regions, vulnerability scanning and build tasks. You pay a daily fee based on the Basic, Standard or Premium tier.","useCases":["Store app images in a private library","Scan container images for security problems","Copy images to other regions"],"pricing":"Daily fee by Basic, Standard or Premium tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/container-registry/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-container-storage","provider":"azure","name":"Azure Container Storage","category":"block-storage","tagline":"Keep data safe for databases and apps running on Kubernetes.","description":"Azure Container Storage gives apps on AKS persistent volumes, which is storage that sticks around even when a container restarts. If you run a database or any app that must remember things, you need this kind of storage. It can sit on top of fast local NVMe drives, Azure Disks or Elastic SAN, so you choose what fits your budget and speed needs. Version 2 is open source, and the service layer itself is free. You only pay for the underlying storage you pick, which makes it a low-cost add-on for stateful workloads.","useCases":["Keep databases running inside containers","Attach persistent disks to stateful apps","Manage storage volumes for container apps"],"pricing":"Free service layer; pay for the underlying storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/container-storage/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-cosmos-db","provider":"azure","name":"Azure Cosmos DB","category":"nosql","also":["vector"],"tagline":"A fast database that keeps your data close to users around the world.","description":"Azure Cosmos DB is a database built to respond in single-digit milliseconds, even when your users are spread across the globe. If you are building an app with customers on several continents, you can copy your data to multiple regions and allow writes in each one. It supports NoSQL, MongoDB, Cassandra, Gremlin and Table APIs, so you can use a style you already know, plus vector search, which helps AI features find similar items. Pick it when global speed matters most. There is a serverless mode and a free tier, and pricing is per request unit.","useCases":["Serve users from several regions at once","Store chat memory for AI assistants","Collect data from connected devices"],"pricing":"Per request unit; free tier available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/cosmos-db/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-documentdb","provider":"azure","name":"Azure DocumentDB","category":"nosql","also":["vector"],"tagline":"A MongoDB-compatible database run for you, built on open source.","description":"Azure DocumentDB is a managed database that works with MongoDB, a popular database that stores data as flexible documents. It was formerly called Azure Cosmos DB for MongoDB (vCore). If your app already uses MongoDB, you can move it here and keep your existing code and habits. It also supports vector search, which lets AI features find similar items. What sets it apart is its base: it runs on the open-source DocumentDB engine, which was donated to the Linux Foundation. You pay per vCore hour, a unit of processing power, plus storage.","useCases":["Move MongoDB apps with few changes","Run MongoDB-compatible databases on Azure","Search data by meaning using vectors"],"pricing":"Per vCore hour and storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/documentdb","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-sql","provider":"azure","name":"Azure SQL","category":"sql","tagline":"Microsoft's family of managed databases, from hands-off to full control.","description":"Azure SQL is a family of three database services that all run on SQL Server, Microsoft's well-known database software. If you are building a new app that needs to store customers, orders or payments in organized tables, you can pick the version that matches how much control you want. Azure SQL Database is the hands-off option, SQL Managed Instance suits moving an existing setup, and SQL Server on VMs (rented virtual servers) gives you full control. Pick this family when your team or tools already speak SQL Server.","useCases":["Modernize older SQL Server systems","Pick the level of control you need","Launch new cloud apps on SQL"],"pricing":"See each service.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-sql/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-sql-database","provider":"azure","name":"Azure SQL Database","category":"sql","tagline":"A fully managed database for your app, with a free monthly allowance.","description":"Azure SQL Database is a database that Microsoft runs for you, so you do not have to handle backups, updates or keeping it online. If you are building a SaaS product (software people subscribe to) that stores customer accounts and billing records, it is a natural home for that data. It tunes itself automatically and supports vector search, which helps AI features find similar items. It also has a serverless option, where capacity adjusts to how much you use it, plus a free offer, so it suits you when you want to start small.","useCases":["Power software-as-a-service back ends","Run business apps on managed SQL","Scale databases up and down automatically"],"pricing":"Per vCore or DTU; free offer available.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-sql/database/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-sql-managed-instance","provider":"azure","name":"Azure SQL Managed Instance","category":"sql","tagline":"A managed database that works like a regular SQL Server, for easy moves.","description":"Azure SQL Managed Instance is a database that Microsoft runs and keeps up to date, designed to behave almost exactly like a traditional SQL Server. If your company has an older app built on SQL Server and you want it in the cloud, this lets you move it with few changes to the app. It supports SQL Agent, a tool for scheduled jobs, and queries that reach across several databases. Pick it over Azure SQL Database when your existing app depends on those features.","useCases":["Move SQL Server apps with few changes","Keep SQL Agent jobs running","Run queries across several databases"],"pricing":"Per vCore hour plus storage; Hybrid Benefit applies.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-sql/managed-instance/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-database-for-postgresql","provider":"azure","name":"Azure Database for PostgreSQL","category":"sql","also":["vector"],"tagline":"Managed Postgres, a popular database, with built-in AI search support.","description":"Azure Database for PostgreSQL is Postgres, a very popular open source database, run for you by Microsoft. If you are building an app that needs to store users, content and settings, it gives you a reliable place to keep them. It supports pgvector, an add-on that stores and searches AI embeddings (numbers that capture meaning), so your app can find similar text or images. It also offers zone-redundant high availability, meaning a standby copy runs in a separate zone, and a free trial for new accounts. Choose it when you want Postgres without managing it.","useCases":["Run app databases with high availability","Power AI search with vector features","Query location-based data"],"pricing":"Per vCore hour plus storage; free trial for new accounts.","status":"ga","url":"https://azure.microsoft.com/en-us/products/postgresql/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-database-for-mysql","provider":"azure","name":"Azure Database for MySQL","category":"sql","tagline":"Managed MySQL, a popular database, for websites and web apps.","description":"Azure Database for MySQL is MySQL, a widely used open source database, run for you by Microsoft. If you are hosting a WordPress site or a PHP app (PHP is a common website language), this is where its posts, users and orders can live. It offers read replicas, which are extra copies that handle lookups to take load off the main database, plus a standby copy in another zone for reliability. Burstable sizes let you start small for light workloads. Pick it when your app already expects MySQL.","useCases":["Host WordPress and PHP websites","Back web apps with managed MySQL","Add read copies for busy sites"],"pricing":"Per vCore hour plus storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/mysql/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-managed-redis","provider":"azure","name":"Azure Managed Redis","category":"kv-cache","also":["vector"],"tagline":"A very fast memory-based database for caching, sessions and AI search.","description":"Azure Managed Redis is a database that keeps data in memory instead of on disk, which makes it extremely quick to read. If your app feels slow because it keeps asking the main database the same questions, you can store the answers here so pages load faster. It also works for remembering who is logged in and for vector search, which finds similar items for AI features. It is built on Redis Enterprise and is the newer replacement for Azure Cache for Redis, so choose it for any new project. A free Azure account trial lets you try it.","useCases":["Cache data for faster responses","Store user sessions in memory","Search vectors held in memory"],"pricing":"Hourly by memory size; free Azure account trial.","status":"ga","url":"https://azure.microsoft.com/en-us/products/managed-redis","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-cache-for-redis","provider":"azure","name":"Azure Cache for Redis","category":"kv-cache","tagline":"The older managed Redis cache, being replaced by Azure Managed Redis.","description":"Azure Cache for Redis is Microsoft's earlier service for keeping frequently used data in fast memory so your app can load it quickly. If you already use it to speed up pages or hold temporary data, it keeps working until the retirement dates below. For anything new, pick Azure Managed Redis, which replaces it. This service is retiring: Enterprise tiers end March 31, 2027, and Basic, Standard and Premium tiers end September 30, 2028.","useCases":["Cache data for older apps","Keep existing caching setups running","Move to Managed Redis before retirement"],"pricing":"Hourly by tier.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/cache/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-database-migration-service","provider":"azure","name":"Azure Database Migration Service","category":"migration","tagline":"Move your existing databases to Azure while keeping downtime low.","description":"Azure Database Migration Service is a tool that moves a database from where it lives now to Azure. If your company keeps customer data in an SQL Server or Oracle database on its own premises, this checks it for problems and then copies it across. It offers online migration, which keeps your app running during the move, and offline migration, which pauses things for a simpler copy. The Standard tier is free, so it is an easy first choice over writing your own copy scripts. Premium is priced per vCore, a unit of computing power.","useCases":["Move SQL Server databases to Azure","Migrate Oracle databases with little downtime","Assess databases before moving them"],"pricing":"Standard tier is free; premium priced per vCore.","status":"ga","url":"https://azure.microsoft.com/en-us/products/database-migration/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-managed-instance-for-apache-cassandra","provider":"azure","name":"Azure Managed Instance for Apache Cassandra","category":"nosql","tagline":"Cassandra databases that Azure scales, patches and backs up for you.","description":"Azure Managed Instance for Apache Cassandra runs Cassandra, a database built to spread huge amounts of data across many machines, and handles the chores for you. If your app collects lots of activity data such as sensor readings or messages and you already use Cassandra, you can move it here without rebuilding. Azure takes care of scaling, patching and backups. It also supports hybrid clusters, which mix machines in Azure with ones elsewhere, so you can migrate gradually. You pay per node hour, meaning per server per hour, as you go.","useCases":["Migrate Cassandra clusters to Azure","Run clusters across on-site and cloud","Automate scaling, patching and backups"],"pricing":"Priced per node hour; pay as you go.","status":"ga","url":"https://azure.microsoft.com/en-us/products/managed-instance-apache-cassandra/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-table-storage","provider":"azure","name":"Table Storage","category":"nosql","tagline":"A cheap, simple place to store lots of small records without a fixed layout.","description":"Table Storage is a low-cost way to save simple records, like items that each have a name and some details, without setting up a rigid structure first. If your app needs to keep user profiles, device readings or small bits of metadata, it is an inexpensive place to put them. It lives inside a storage account, the same place Azure keeps your files. If you later need to serve users worldwide or search in more ways, Cosmos DB for Table is the upgrade path. You pay per gigabyte stored plus per read or write.","useCases":["Store simple user profile records","Save device readings at low cost","Keep app metadata in cheap storage"],"pricing":"Per GB stored plus per transaction.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/tables/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-devops","provider":"azure","name":"Azure DevOps","category":"devtools","tagline":"Plan work, store code, and ship updates automatically, all in one place.","description":"Azure DevOps is a bundle of tools for teams building software. Boards track tasks, Repos store your code using Git (the standard way to track code changes), Pipelines automatically test and release your app, and Test Plans and Artifacts cover testing and shared packages. If you have a small team shipping a product, you can plan, code and publish without stitching together several separate tools. It supports agile planning and AI-assisted development, and it is free for small teams, with paid charges per user and per parallel job beyond that.","useCases":["Plan team work and sprints","Host Git code repositories","Run build and release pipelines"],"pricing":"Free for small teams; paid per user and per parallel job beyond that.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-devtest-labs","provider":"azure","name":"Azure DevTest Labs","category":"devtools","tagline":"Spin up test environments from templates, with spending limits built in.","description":"Azure DevTest Labs lets a team create practice and testing setups from reusable templates, so everyone starts from the same baseline. If you are building an app and want a safe copy of your environment to try changes, or you run a training class that needs identical computers for each student, this fits well. It adds quotas, which cap how much each person can create, and auto-shutdown, which turns machines off on a schedule so you stop paying for idle ones. The service itself is free, and you pay only for the resources you create.","useCases":["Give teams ready-made dev environments","Set quotas and automatic shutdowns","Set up training labs"],"pricing":"Free service; pay for the resources created.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devtest-lab/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-pipelines","provider":"azure","name":"Azure Pipelines","category":"devtools","tagline":"Automatically build, test and ship your code to any cloud.","description":"Azure Pipelines is a service that automatically builds, tests and publishes your code every time you make a change. If you have a web app, you can set it up so each update gets checked for mistakes and then goes live without you doing it by hand. It works with any programming language and any cloud, not just Microsoft's, and you describe the steps in YAML, a simple text format for settings. You can run jobs on machines Microsoft provides or on your own. You get free parallel jobs and minutes to start, then pay for more parallel jobs.","useCases":["Build and test code automatically","Deploy apps to any cloud","Run jobs on Microsoft or own machines"],"pricing":"Free parallel jobs and minutes, then paid parallel jobs.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/pipelines/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-visual-studio","provider":"azure","name":"Visual Studio","category":"devtools","tagline":"A full-featured Microsoft app for writing, testing and publishing code.","description":"Visual Studio is Microsoft's full code-writing app, called an IDE (a workspace that bundles an editor with testing and debugging tools). If you're building a business app with .NET, Microsoft's framework for making apps, it lets you hunt down bugs and publish to Azure from the same window. It includes GitHub Copilot, an AI coding helper, and runs as a Windows desktop app. Pick it over a lighter editor when you want everything built in for .NET or C++ projects. The Community edition is free, while Professional and Enterprise are paid.","useCases":["Write .NET apps on Windows","Debug apps step by step","Publish projects straight to Azure"],"pricing":"Free Community edition; Professional and Enterprise paid.","status":"ga","url":"https://azure.microsoft.com/en-us/products/visual-studio/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-visual-studio-code","provider":"azure","name":"Visual Studio Code","category":"devtools","tagline":"A free code editor you can customize with add-ons.","description":"Visual Studio Code is a free app for writing and editing code. If you're tweaking an AI-generated project or building from scratch, it's a simple place to open your files, make changes and install add-ons for the languages and tools you use. Its Azure add-ons include emulators, which are pretend local copies of Azure Storage and Cosmos DB, so you can try things out without touching the real services. Pick it over Visual Studio when you want something lighter that runs on different kinds of computers and also supports remote development.","useCases":["Edit code in a free editor","Test storage locally with emulators","Code on remote machines"],"pricing":"Free.","status":"ga","url":"https://azure.microsoft.com/en-us/products/visual-studio-code/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-app-testing","provider":"azure","name":"Azure App Testing","category":"devtools","tagline":"Test how your app handles crowds and how it works in different browsers.","description":"Azure App Testing is a managed service for checking that your app holds up under heavy traffic and works in different web browsers. If you're about to launch, you can simulate lots of visitors at once with JMeter and Locust, two open-source load testing tools, or run Playwright tests that click through your site like a real person would. It combines Azure Load Testing with Playwright Workspaces, so both kinds of testing live in one place. It replaced Microsoft Playwright Testing and the old Azure Load Testing name. You pay as you go, based on virtual user hours and test minutes.","useCases":["Run load tests with many users","Test sites across several browsers","Run end-to-end browser tests"],"pricing":"Pay as you go by virtual user hours and test minutes.","status":"ga","url":"https://azure.microsoft.com/en-us/products/app-testing/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-dev-box","provider":"azure","name":"Microsoft Dev Box","category":"devtools","tagline":"Ready-to-code workstations that live in the cloud.","description":"Microsoft Dev Box gives each developer a computer that lives in the cloud, already set up with the tools they need. If you bring on a freelancer or a new teammate, they can start coding the same day instead of spending days installing software. Workstations are organized by project and managed through Intune, Microsoft's tool for controlling company devices. Pick it over setting up laptops by hand when you want the same setup for everyone, or a specialized machine for certain work. You pay per hour for the computer, plus storage.","useCases":["Onboard new developers with ready workstations","Give developers secure cloud computers","Set up specialized workstations for projects"],"pricing":"Pay per hour for compute plus storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/dev-box/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-deployment-environments","provider":"azure","name":"Azure Deployment Environments","category":"devtools","tagline":"Let developers spin up approved app setups from ready-made templates.","description":"Azure Deployment Environments lets a team create complete app setups, including the servers and databases behind them, from templates someone has approved in advance. If several developers each need their own test copy of your app, they can create one on their own without asking for help or guessing at settings. Templates are written as IaC (infrastructure as code, meaning setup instructions saved in files) and organized by project. Choose it over building each setup by hand when you want every environment to match. There is no fee for the service itself, and you pay only for the resources it creates.","useCases":["Let developers spin up sandboxes","Standardize environments from approved templates","Give teams self-service infrastructure"],"pricing":"No service fee; pay for deployed resources.","status":"ga","url":"https://azure.microsoft.com/en-us/products/deployment-environments/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-artifacts","provider":"azure","name":"Azure Artifacts","category":"devtools","tagline":"A private shelf for the code packages your team builds and reuses.","description":"Azure Artifacts is a place to store and share packages, which are reusable bundles of code, in formats like npm, NuGet, Maven and Python. If you've written a helper that several of your apps use, you can publish it privately and let each app pull it in. You can create feeds (collections of packages) from public and private sources, and add package sharing to your pipelines, which are automated build steps. It also supports upstream caching, which means saving copies of packages from public sources. You get 2 GiB free, then pay per GiB.","useCases":["Host private npm and NuGet packages","Share Python and Maven packages","Cache public packages upstream"],"pricing":"Free 2 GiB, then billed per GiB.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/artifacts/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-boards","provider":"azure","name":"Azure Boards","category":"devtools","tagline":"Track tasks, bugs and sprints on boards your whole team can see.","description":"Azure Boards is a tool for planning and tracking your team's work. If you're juggling a long list of features and bugs, you can lay them out on Scrum or Kanban boards (card-based views that move tasks from to do to done), plan sprints, which are short work cycles, and check progress on dashboards. It works with any agile method, so you aren't forced into one way of working. It's free for up to five users, which suits a small founding team, with per-user pricing after that.","useCases":["Track bugs and feature requests","Plan sprints with backlogs","Use Scrum or Kanban boards"],"pricing":"Free for up to five users; per-user pricing after.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/boards/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-repos","provider":"azure","name":"Azure Repos","category":"devtools","tagline":"A private home for your code, with history and pull request reviews.","description":"Azure Repos stores your code in Git, the most common tool for tracking every change made to a project. If you want teammates to check changes before they go live, they can use pull requests, which are proposals to add changes that others can comment on and approve. You get unlimited private repositories, meaning code storage spaces only your team can see. It was formerly part of Visual Studio Team Services. It's free for small teams and included with Azure DevOps, Microsoft's suite of development tools.","useCases":["Host private code in Git","Review code changes before merging","Track changes through pull requests"],"pricing":"Free for small teams; included with Azure DevOps.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/repos/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-test-plans","provider":"azure","name":"Azure Test Plans","category":"devtools","tagline":"Organize hands-on testing where people click through your app by hand.","description":"Azure Test Plans is a browser-based tool for managing testing done by people instead of automatic scripts. If you're about to release a new checkout flow, you can write down the steps for testers to follow and record what passes or fails. It also supports exploratory testing, where a tester pokes around freely to find problems nobody planned for. Choose it when human judgment matters more than automated checks. It's a paid per-user add-on to Azure DevOps, Microsoft's suite of development tools.","useCases":["Run manual tests in a browser","Run exploratory testing sessions","Manage test cases in one place"],"pricing":"Paid per-user add-on to Azure DevOps.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/test-plans/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-github-advanced-security-for-azure-devops","provider":"azure","name":"GitHub Advanced Security for Azure DevOps","category":"threat-intel","tagline":"Scan your code for leaked passwords, risky packages and security bugs.","description":"GitHub Advanced Security for Azure DevOps brings GitHub's security checks to code stored in Azure Repos. If you or an AI tool accidentally paste an API key (a secret code that gives access to a service) into your code, secret scanning can catch it. Dependency scanning flags risky outside packages your app relies on, and code scanning uses static analysis, which reads your code without running it, to spot security flaws. Choose it when your code already lives in Azure Repos. It's billed per active committer, meaning each person who adds code.","useCases":["Scan code for security flaws","Find passwords left in code","Flag risky third-party dependencies"],"pricing":"Billed per active committer.","status":"ga","url":"https://azure.microsoft.com/en-us/products/devops/github-advanced-security","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-github-enterprise","provider":"azure","name":"GitHub Enterprise","category":"devtools","tagline":"GitHub for larger teams, bought through your Azure agreement.","description":"GitHub Enterprise is GitHub's plan for companies, an AI-powered platform for storing code and working on it together. If your company already pays Microsoft through an Azure agreement, you can buy it there instead of setting up a separate contract. It adds enterprise controls, meaning extra settings for managing who can do what. Pick it over the regular GitHub plans when you need that company-level oversight. Pricing is per user per month.","useCases":["Host company code on GitHub","Collaborate on code across teams","Buy GitHub through an Azure agreement"],"pricing":"Per user per month.","status":"ga","url":"https://azure.microsoft.com/en-us/products/github","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-github-copilot","provider":"azure","name":"GitHub Copilot","category":"devtools","tagline":"An AI pair programmer that writes and explains code as you type.","description":"GitHub Copilot is an AI assistant that suggests code, answers questions and explains what existing code does, right inside your code editor and on GitHub. If you get stuck on a bug or need tests written for a new feature, you can ask it in plain language. It also has an agent mode, which lets it work through multi-step tasks instead of offering one suggestion at a time. Pick it when you want help that understands your own project rather than generic answers. It is billed per user per month, with a limited free plan.","useCases":["Generate code from written prompts","Write tests for existing code","Explain unfamiliar code in plain terms"],"pricing":"Per user per month with a limited free plan.","status":"ga","url":"https://azure.microsoft.com/en-us/products/github/Copilot","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-monitor","provider":"azure","name":"Azure Monitor","category":"observability","tagline":"See how your app and servers are doing, and get alerted when something breaks.","description":"Azure Monitor collects numbers, logs (a running diary of what your software did) and traces (the path a request took through your app) from your app and the machines it runs on. If your checkout page suddenly gets slow, you can see where the time went and set an alert to notify you next time. It includes Application Insights, Log Analytics and managed Prometheus (a popular open-source tool for app measurements), plus dashboards. Pick it when your app already runs on Azure and you want monitoring built in. You pay per GB of data collected, and free allowances apply.","useCases":["Track app performance and errors","Search and analyze application logs","Send alerts when things break"],"pricing":"Pay per GB of data ingested; free allowances apply.","status":"ga","url":"https://azure.microsoft.com/en-us/products/monitor/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-managed-grafana","provider":"azure","name":"Azure Managed Grafana","category":"observability","tagline":"Ready-made dashboards that turn your Azure data into charts.","description":"Azure Managed Grafana is Grafana, a popular dashboard tool, run for you by Azure so you don't have to set it up or maintain it. If you want one screen showing signups, errors and server health, you can pull data from several sources into charts, logs and alerts. It is the Grafana Enterprise edition, and it works well for showing Prometheus metrics (a common format for app measurements). Pick it over hosting Grafana yourself when you would rather skip the upkeep. Pricing is hourly per instance.","useCases":["Build dashboards from monitoring data","Chart Prometheus metrics visually","Combine data from several sources"],"pricing":"Hourly per instance.","status":"ga","url":"https://azure.microsoft.com/en-us/products/managed-grafana/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-arc","provider":"azure","name":"Azure Arc","category":"management","tagline":"Manage servers and apps outside Azure from the same Azure control panel.","description":"Azure Arc lets you manage machines and apps that live outside Azure, such as in your own office or on another cloud, from the Azure portal. It covers servers, Kubernetes clusters (groups of machines that run packaged apps) and data services. If your setup is a mix of places, Arc gives you one spot for security rules and oversight. Pick it when moving everything to Azure isn't realistic but you still want consistent management. The core control layer is free, though some add-on services are billed.","useCases":["Manage servers and clusters anywhere","Govern Kubernetes clusters outside Azure","Apply security policies across environments"],"pricing":"Core control plane is free; some services are billed.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-arc/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-expressroute","provider":"azure","name":"Azure ExpressRoute","category":"networking","tagline":"A private line from your network to Azure that skips the public internet.","description":"Azure ExpressRoute is a private, dedicated connection between your own network and Azure, so your traffic never travels over the public internet. If your company moves large amounts of data between its offices or data center and the cloud, it offers steadier speeds and higher reliability than a normal connection. It supports private connections including MPLS, a type of private business network. Pick it over a regular internet link when predictable performance matters more than simplicity. You pay a monthly fee per connection plus outbound data.","useCases":["Connect office network privately to Azure","Move large data sets reliably","Avoid the public internet for steady speed"],"pricing":"Monthly circuit fee plus outbound data.","status":"ga","url":"https://azure.microsoft.com/en-us/products/expressroute/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-iot-edge","provider":"azure","name":"Azure IoT Edge","category":"iot","tagline":"Run your cloud code directly on devices, even with a weak connection.","description":"Azure IoT Edge lets you run software on physical devices, such as a camera or a factory sensor, instead of only in the cloud. If your product analyzes video on-site, you can package AI and custom logic as containers (self-contained bundles of software) and send them to those devices. You manage them remotely through IoT Hub, Azure's service for connecting devices. That means data can be processed locally, even when the device is offline. The runtime is free, and you pay for IoT Hub.","useCases":["Run custom logic on devices","Keep processing working when offline","Deploy AI models to devices"],"pricing":"Free runtime; pay for IoT Hub.","status":"ga","url":"https://azure.microsoft.com/en-us/products/iot-edge/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-sentinel","provider":"azure","name":"Microsoft Sentinel","category":"threat-intel","tagline":"Spots security threats across your systems and helps you respond fast.","description":"Microsoft Sentinel is a security tool that gathers security data from across your clouds and flags suspicious activity. It is known as a SIEM, a system that collects and analyzes security alerts, and it uses rules to detect threats and playbooks (automated response steps) to react. If your product handles customer data, it can help your team notice and investigate unusual activity. Pick it when you want one view across several clouds. It is now integrated into the Defender portal. Pricing is per GB ingested, with commitment tiers available.","useCases":["Detect threats across multiple clouds","Respond to security incidents automatically","Analyze security logs in one place"],"pricing":"Pay per GB ingested; commitment tiers available.","status":"ga","url":"https://www.microsoft.com/en-us/security/business/siem-and-xdr/microsoft-sentinel/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-defender-for-cloud","provider":"azure","name":"Microsoft Defender for Cloud","category":"threat-intel","also":["management"],"tagline":"Checks your cloud setup for security gaps and protects what runs on it.","description":"Microsoft Defender for Cloud reviews how your cloud is set up and points out security weak spots, which is called CSPM (cloud security posture management). It also protects the servers and containers (packaged apps) you run. If a setting is left open by mistake, it can flag the problem before someone else finds it. Unlike many Azure tools, it covers AWS and GCP resources too, so it suits teams using more than one cloud. Basic posture checks are free, and paid plans are priced by resource.","useCases":["Check cloud settings for security gaps","Protect servers and containers","Secure AWS and GCP resources too"],"pricing":"Free foundational CSPM; paid plans by resource.","status":"ga","url":"https://www.microsoft.com/en-us/security/business/cloud-security/microsoft-defender-cloud/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-defender-external-attack-surface-management","provider":"azure","name":"Defender External Attack Surface Management","category":"threat-intel","tagline":"Finds your internet-facing sites and servers, including forgotten ones.","description":"Defender External Attack Surface Management continuously maps what your organization exposes to the internet, then flags the risky items. If a teammate once launched a test website and forgot about it, this is how you would find it. That kind of hidden, unapproved technology is often called shadow IT. Pick it when your company has grown across many domains and tools and nobody has a complete list. It is billed per managed asset per day.","useCases":["Find forgotten internet-facing assets","Discover unknown internet-exposed systems","Monitor exposed assets continuously"],"pricing":"Billed per managed asset per day.","status":"ga","url":"https://azure.microsoft.com/en-us/products/defender-external-attack-surface-management/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-local","provider":"azure","name":"Azure Local","category":"vms","tagline":"Azure's tools running on your own hardware in your own location.","description":"Azure Local, formerly called Azure Stack HCI, runs virtual machines (software-based computers) and AKS, Azure's managed Kubernetes service for packaged apps, on pre-approved hardware in your own building. You still manage it through Azure. If a store, factory or government office needs apps running on-site, this keeps them there. It also suits cases with strict rules about where data sits. Pick it over the regular Azure cloud when the computers must physically be with you. Pricing is per physical core per month.","useCases":["Run virtual machines on-premises","Operate edge sites locally","Support sovereign deployments on own hardware"],"pricing":"Per physical core per month.","status":"ga","url":"https://azure.microsoft.com/en-us/products/local/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-stack-hub","provider":"azure","name":"Azure Stack Hub","category":"vms","tagline":"A ready-built Azure setup that lives in your own data center.","description":"Azure Stack Hub is a complete package of hardware with Azure software already installed, which you place in your own data center. If you work in a regulated field or somewhere without reliable internet, it lets you run Azure services even when disconnected from the cloud. It also helps when rules say your data must stay in a specific place, known as data residency. Pick it when you want a pre-built system delivered as one unit. Cost is the hardware plus usage-based service fees.","useCases":["Run cloud services in own datacenter","Operate disconnected from the internet","Keep data within required regions"],"pricing":"Hardware cost plus usage-based service fees.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-stack/hub/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-stack-edge","provider":"azure","name":"Azure Stack Edge","category":"iot","tagline":"A managed Azure device that brings storage and AI to your location.","description":"Azure Stack Edge is a piece of hardware provided as a service, which brings Azure's computing, storage and AI to wherever you work. If you run cameras in a warehouse and want to spot problems instantly, it can run AI there (called inference) without sending everything to the cloud. It also handles other data processing locally. Pick it when you want a ready-made box rather than building your own setup at the edge. You pay a monthly device subscription.","useCases":["Run AI inference at the edge","Process data on local devices","Manage edge hardware as a subscription"],"pricing":"Monthly device subscription.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-stack/edge/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-operator-service-manager","provider":"azure","name":"Azure Operator Service Manager","category":"specialized","tagline":"Roll out and update telecom network software from one place.","description":"Azure Operator Service Manager helps phone and internet carriers deploy, upgrade and manage the software that runs their networks, using one shared set of automation tools. If you work at a carrier and need to roll out 5G network functions (the software pieces that handle calls and data), it keeps releases and updates consistent instead of handled by hand. Pick it when you manage many telecom services over their whole life and want one orchestration layer. Most app builders will never need it. Pricing is usage-based, so check the product page.","useCases":["Deploy 5G network functions","Upgrade telecom services with automation","Manage telecom services lifecycle"],"pricing":"Usage-based; see the product page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/operator-service-manager/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-operator-nexus","provider":"azure","name":"Azure Operator Nexus","category":"specialized","tagline":"A cloud platform built for running mobile networks.","description":"Azure Operator Nexus is a hybrid cloud platform, meaning it spans your own sites and Azure, built for the apps that keep mobile networks running. If you are a carrier hosting the core of a 5G network, it runs both containerized software (small packaged apps) and virtualized network functions (network gear turned into software). It is designed for mission-critical telecom workloads, which sets it apart from general-purpose cloud servers. It is a specialist product for telecom companies, and pricing comes through sales.","useCases":["Host 5G core network software","Run telecom workloads on hybrid cloud","Run mission-critical mobile network apps"],"pricing":"Contact sales.","status":"ga","url":"https://azure.microsoft.com/en-us/products/operator-nexus/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-network-function-manager","provider":"azure","name":"Azure Network Function Manager","category":"networking","tagline":"Run telecom network software on devices at your own sites.","description":"Azure Network Function Manager lets you deploy and manage network software on edge devices, which are machines sitting at your own location instead of in a data center. If you want a private 5G or LTE network at a factory, or SD-WAN (smart routing between office networks), you can pick the software from Azure Marketplace and manage it from the cloud. It is a good fit when network features must run on site but you want one place to control them. Pricing is usage-based, so see the product page.","useCases":["Deploy private 5G at sites","Run SD-WAN at edge locations","Manage network functions on edge devices"],"pricing":"Usage-based; see the product page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-network-function-manager/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-storage-mover","provider":"azure","name":"Azure Storage Mover","category":"migration","tagline":"Move your files into Azure Storage safely, with no service fee.","description":"Azure Storage Mover is a managed service that copies your files from your own servers or another cloud into Azure Storage. If your team keeps documents or media on a network drive (a NAS) and wants them in the cloud, you set up the transfer and let it run. It suits large file moves where you want safe, repeatable copying without writing scripts. There is no charge for the service itself, and you only pay for the storage and transactions you use.","useCases":["Move files from on-premises servers","Migrate NAS storage to Azure","Transfer files between clouds"],"pricing":"No service charge; pay for storage and transactions.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage-mover","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-entra-id","provider":"azure","name":"Microsoft Entra ID","category":"identity","tagline":"Let your team sign in once and keep company accounts secure.","description":"Microsoft Entra ID, formerly Azure Active Directory, manages who can log in to your tools and apps. If your team uses many apps, it gives them single sign-on (one login for everything) and MFA (a second check, like a phone prompt). Conditional access rules can also block risky sign-ins. Pick it when you need to control access for employees, especially if you already use Microsoft 365. There is a free tier, and paid P1 and P2 plans are priced per user.","useCases":["Give staff single sign-on","Require MFA and conditional access","Register app identities"],"pricing":"Free tier; paid P1 and P2 per user.","status":"ga","url":"https://www.microsoft.com/en-us/security/business/identity-access/microsoft-entra-id","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-entra-domain-services","provider":"azure","name":"Microsoft Entra Domain Services","category":"identity","tagline":"Old-style company logins in the cloud, without running the servers.","description":"Microsoft Entra Domain Services, formerly Azure AD Domain Services, gives you classic company login features such as domain join, LDAP, Kerberos and group policy. These are older standards many business apps expect. If you are moving an older app to the cloud that needs them, this saves you from setting up and maintaining your own domain controllers (the servers that handle logins). Pick it over Microsoft Entra ID when an app cannot work without those older methods. It is billed hourly by plan size.","useCases":["Run legacy apps that need domains","Apply group policy without domain controllers","Support LDAP and Kerberos logins"],"pricing":"Hourly by SKU.","status":"ga","url":"https://azure.microsoft.com/en-us/products/microsoft-entra-ds","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-entra-external-id","provider":"azure","name":"Microsoft Entra External ID","category":"identity","tagline":"Add sign-up and login for your customers and partners.","description":"Microsoft Entra External ID handles sign-up and sign-in for the people who use your product, such as customers and partners. If you are building an app where users create accounts, you can offer a login screen styled with your own branding without building the security yourself. It replaces Azure AD B2C for new customers, so it is the one to choose today. It is free for your first 50,000 monthly active users, meaning people who sign in during a month.","useCases":["Sign up and sign in customers","Give partners access to apps","Brand customer sign-in pages"],"pricing":"Free for the first 50,000 monthly active users.","status":"ga","url":"https://www.microsoft.com/en-us/security/business/identity-access/microsoft-entra-external-id","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-active-directory-b2c","provider":"azure","name":"Azure Active Directory B2C","category":"identity","tagline":"Older customer login service, closed to new customers.","description":"Azure Active Directory B2C is an older service that handled sign-up and login for an app's customers. If you already run an app on it, it keeps working, and existing tenants are billed per monthly active user, meaning each person who signs in during a month. New projects should look at Microsoft Entra External ID, which replaces it. Azure AD B2C is no longer sold to new customers as of May 1, 2025.","useCases":["Keep existing customer sign-in running","Serve current customers' login pages","Move customers to Entra External ID"],"pricing":"Per monthly active user for existing tenants.","status":"closed","url":"https://learn.microsoft.com/en-us/azure/active-directory-b2c/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-api-management","provider":"azure","name":"API Management","category":"devtools","tagline":"One front door to publish, protect and track your APIs.","description":"API Management is a front door for your APIs, the connections that let apps and services talk to each other. If you offer an API to customers or partners, it lets you publish it, control who gets in, limit traffic and see how it is used. It also includes a developer portal where outside developers can read the docs, and rules for managing traffic to AI models. Pick it when you need control over many APIs in one place. Pricing is hourly by tier, or pay per call on the consumption tier.","useCases":["Publish and secure APIs","Give developers an API portal","Apply policies to AI model traffic"],"pricing":"Hourly by tier; consumption tier is pay per call.","status":"ga","url":"https://azure.microsoft.com/en-us/products/api-management/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-health-data-services","provider":"azure","name":"Azure Health Data Services","category":"specialized","tagline":"Store and share patient records and medical images in the cloud.","description":"Azure Health Data Services is a set of services made for protected health information, the sensitive patient data that has strict privacy rules. If you are building an app that reads patient records or stores medical images, it supports FHIR (a common format for health records), DICOM (the standard for medical images) and MedTech for device data. Pick it over a general database when you need those health standards built in. Pricing is usage-based and depends on which service you use.","useCases":["Store health records through APIs","Manage medical imaging files","Handle protected patient information securely"],"pricing":"Usage-based by service.","status":"ga","url":"https://azure.microsoft.com/en-us/products/health-data-services/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-event-grid","provider":"azure","name":"Event Grid","category":"queues","tagline":"Pass messages between your apps and devices when things happen.","description":"Event Grid is a managed service that sends a notice from one place to another when something happens. If your app should react when a file is uploaded, such as by resizing a photo, Event Grid delivers that signal to the right part of your system. It also has an MQTT broker, a messaging hub that smart devices use to report in. Pick it when you want parts of your app to respond to events without constant checking. It bills per operation, and the first 100,000 each month are free.","useCases":["React when events happen in apps","Route events between services automatically","Send messages between smart devices"],"pricing":"Billed per operation; first 100,000 per month free.","status":"ga","url":"https://azure.microsoft.com/en-us/products/event-grid/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-logic-apps","provider":"azure","name":"Logic Apps","category":"workflows","tagline":"Connect your tools and automate tasks with little or no code.","description":"Logic Apps lets you build automated workflows by linking steps together visually, with little or no code. If you want a new form submission to create a customer record, send an email and ask a manager for approval, you can build that from ready-made connectors, which are plug-ins for hundreds of services. It also supports AI workflows and multi-agent automation. Pick it when you want to connect systems quickly instead of writing custom code. Pricing is per action run, or a fixed plan price for Standard.","useCases":["Connect many apps without coding","Automate approval steps for teams","Run multi-step business processes"],"pricing":"Per action execution, or fixed plan pricing for Standard.","status":"ga","url":"https://azure.microsoft.com/en-us/products/logic-apps/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-service-bus","provider":"azure","name":"Service Bus","category":"queues","tagline":"Pass messages between parts of your app so none get lost.","description":"Service Bus is a message broker, a middleman that holds messages until the part of your app that needs them is ready. If your store takes orders, you can drop each one into a queue (a waiting line for messages) and let a separate service handle payment and shipping at its own pace. It can also send one message to many listeners through topics, keep messages in order, and set aside ones that keep failing (called dead-lettering). Pick it over direct calls between services when losing an order would hurt. You pay as you go by tier, and Basic is the cheapest.","useCases":["Queue orders for reliable processing","Let services work without waiting","Hold failed messages for review"],"pricing":"Pay as you go by tier; Basic is cheapest.","status":"ga","url":"https://azure.microsoft.com/en-us/products/service-bus/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-web-pubsub","provider":"azure","name":"Azure Web PubSub","category":"realtime","tagline":"Push live updates to many users' browsers at the same moment.","description":"Azure Web PubSub is a hosted service for live, two-way messaging between your app and your users' browsers, using WebSockets (a standard way to keep a connection open). If you are building live chat or a dashboard that updates without anyone refreshing, this handles the connections for you. It is serverless, which means you don't run or size the servers yourself. Pick it when you want publish and subscribe style messaging, where people join a channel and receive what gets sent to it. You pay per unit per day.","useCases":["Build live chat in web apps","Push live dashboard updates","Send messages both ways instantly"],"pricing":"Per unit per day.","status":"ga","url":"https://azure.microsoft.com/en-us/products/web-pubsub/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-signalr-service","provider":"azure","name":"Azure SignalR Service","category":"realtime","tagline":"Add live updates and notifications to your web app without the plumbing.","description":"Azure SignalR Service keeps web pages updated in real time by managing all the open connections between your app and its visitors. If you want a notification to pop up the moment something happens, or a page to change without a refresh, you turn this on instead of building that machinery yourself. It is made for ASP.NET, Microsoft's web framework, so it fits best if your app already uses it. There is a free tier to try it, and the Standard tier is priced per unit per day.","useCases":["Add live updates to web apps","Send instant notifications to users","Run real-time features without managing connections"],"pricing":"Free tier; Standard per unit per day.","status":"ga","url":"https://azure.microsoft.com/en-us/products/signalr-service/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-fluid-relay","provider":"azure","name":"Azure Fluid Relay","category":"realtime","tagline":"Let several people edit the same thing together, live, in your app.","description":"Azure Fluid Relay is a managed service that keeps shared sessions in sync with very little delay. If you are building a shared whiteboard or a document that several people edit at once, it passes everyone's changes back and forth so each person sees the same thing. It is made for apps built with the Fluid Framework, Microsoft's toolkit for collaborative features, so it is not a general chat or messaging tool. Pick it when real-time co-editing is the heart of your product. Pricing is based on usage.","useCases":["Let several people co-edit documents","Power shared whiteboard canvases","Keep collaborative sessions in sync"],"pricing":"Usage-based; see the pricing page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/fluid-relay/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-digital-twins","provider":"azure","name":"Azure Digital Twins","category":"iot","tagline":"Build a live digital copy of a building, factory or other space.","description":"Azure Digital Twins lets you create a digital model of a real place and how its parts connect, such as rooms, machines and the sensors inside them. If you manage smart buildings or monitor a factory, you can see what is happening across the whole space instead of device by device. The model is a graph, which is a map of things and their relationships. Pick it when the connections between your devices matter as much as the readings themselves. You pay per operation, message and query unit.","useCases":["Model buildings as digital replicas","Monitor factories through digital models","Map relationships between devices and rooms"],"pricing":"Per operation, message and query unit.","status":"ga","url":"https://azure.microsoft.com/en-us/products/digital-twins/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-iot-central","provider":"azure","name":"Azure IoT Central","category":"iot","tagline":"A ready-made app for connecting and watching your smart devices.","description":"Azure IoT Central is a managed app for connecting internet-enabled devices (IoT, short for Internet of Things) and seeing what they report. If you want to test a connected product idea, it gives you ready-made screens and an API, so you don't have to build a dashboard first. Pick it over assembling the pieces yourself when speed matters more than total control, such as a proof of concept or a simple device dashboard. You pay per device per month, which makes costs easy to predict.","useCases":["Launch an IoT pilot quickly","Build device dashboards without coding","Manage devices from a ready-made app"],"pricing":"Per device per month.","status":"ga","url":"https://azure.microsoft.com/en-us/products/iot-central/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-iot-hub","provider":"azure","name":"Azure IoT Hub","category":"iot","tagline":"Connect to your devices, receive their data and send them commands.","description":"Azure IoT Hub is the central meeting point between your app and your connected devices (IoT, short for Internet of Things). If you have sensors sending readings, called telemetry, or gadgets you need to control from afar, they all talk to the hub securely in both directions. It also helps you set up and configure new devices as you add them. Pick it when you want to build your own device system rather than use a ready-made one. There is a free tier, and paid plans depend on tier and message volume.","useCases":["Send device data to the cloud","Send commands to remote devices","Onboard and configure new devices"],"pricing":"Free tier; paid by tier and message volume.","status":"ga","url":"https://azure.microsoft.com/en-us/products/iot-hub/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-iot-operations","provider":"azure","name":"Azure IoT Operations","category":"iot","tagline":"Gather data from equipment on site and send it to the cloud.","description":"Azure IoT Operations runs on your own premises, such as a factory floor, and collects data from industrial equipment before sending it to the cloud. It runs on Kubernetes (a system for running software across machines) connected to Azure through Azure Arc, and it speaks MQTT and OPC UA, two common languages that machines use. If you need to analyze factory data close to where it is made, this is built for that. It is aimed at industrial setups, so most app builders won't need it. Pricing is usage-based per core.","useCases":["Collect data from factory equipment","Analyze machine data at the edge","Send equipment data to the cloud"],"pricing":"Usage-based per core; see the pricing page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/iot-operations","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-maps","provider":"azure","name":"Azure Maps","category":"specialized","tagline":"Add maps, directions and address search to your app.","description":"Azure Maps gives your app map features through simple building blocks: search, routing, traffic, weather and time zones. If you are building a store locator, you can turn an address into a spot on the map (called geocoding) and show the nearest shop. A delivery product can use the same service to plan routes for a fleet. It works in both web and mobile apps. You pay per transaction, and there is a free monthly allowance, so small projects can start without a bill.","useCases":["Add store locator maps","Plan driving routes for fleets","Look up addresses and places"],"pricing":"Pay per transaction; free monthly allowance.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-maps/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-notification-hubs","provider":"azure","name":"Notification Hubs","category":"messaging","tagline":"Send push notifications to phones and devices from one place.","description":"Notification Hubs sends push notifications, the alerts that appear on a phone even when your app is closed, from any back end. If you have an app on both iPhone and Android, you send one message here instead of connecting to each platform separately. It also supports Windows and Kindle. Pick it when you need to reach several kinds of devices or broadcast an alert to many people at once. There is a free tier, and paid use is priced per million pushes.","useCases":["Send push alerts to phone apps","Broadcast alerts to all devices","Reach iPhone and Android users together"],"pricing":"Free tier; paid per million pushes.","status":"ga","url":"https://azure.microsoft.com/en-us/products/notification-hubs/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-windows-for-iot","provider":"azure","name":"Windows for IoT","category":"iot","tagline":"Windows built for kiosks, machines and other single-purpose devices.","description":"Windows for IoT is a version of Windows made for dedicated devices, such as kiosks and industrial PCs, rather than everyday laptops. If you are building a self-service screen or a machine controller, Windows IoT Enterprise gives you a familiar system with long-term support and enterprise-grade security. Pick it when your device needs to run Windows software. It is licensed through device makers, so you usually get it by buying hardware that already includes it.","useCases":["Run kiosks on Windows","Power industrial PCs and embedded devices","Get long-term support for devices"],"pricing":"Licensed through device makers.","status":"ga","url":"https://azure.microsoft.com/en-us/products/windows-iot/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-sphere","provider":"azure","name":"Azure Sphere","category":"iot","tagline":"Secure chip and cloud for connected devices, now retiring.","description":"Azure Sphere is a hardware and cloud security platform for connected devices, built around a chip called the MT3620. It was made to keep small devices protected and updated, and it only matters today if you already have devices in the field. It is not an option for new projects, and pricing applies to existing customers only. Microsoft announced its retirement on March 20, 2026. The MT3620 reaches end of life on July 31, 2026, and operating system and security updates end on July 31, 2031.","useCases":["Keep already deployed devices running","Get security updates until July 2031","Plan a move off the retiring platform"],"pricing":"Existing customers only.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/azure-sphere/","source":"https://learn.microsoft.com/en-us/azure-sphere/product-overview/retirement"},{"id":"azure-automation","provider":"azure","name":"Automation","category":"management","tagline":"Run repeat chores in Azure on a schedule, using small scripts.","description":"Automation lets you write small scripts, called runbooks, that carry out chores in Azure for you, in PowerShell or Python (two common scripting languages). If your app needs test servers shut down every night to save money, or updates installed every weekend, you can set a runbook to do it on a schedule. It also includes tools for installing updates and keeping settings consistent. Pick it when you want Azure itself to run your routine tasks without you logging in. You get some free minutes each month, then pay for each minute a job runs.","useCases":["Run scheduled scripts automatically","Automate software updates on servers","Handle repetitive tasks without supervision"],"pricing":"Free monthly minutes, then per job runtime minute.","status":"ga","url":"https://azure.microsoft.com/en-us/products/automation/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-advisor","provider":"azure","name":"Azure Advisor","category":"management","tagline":"Get free tips on how to cut costs and tighten up your Azure setup.","description":"Azure Advisor looks at how you have set up and used Azure, then gives you personalized advice. If your app has grown quickly and you suspect you are paying for things you no longer need, it can point out where to save money. It also flags security gaps, reliability risks and ways to make things faster. Pick it as a free first stop for a health check before you hire a consultant or dig through settings by hand. It costs nothing to use.","useCases":["Find ways to cut cloud costs","Spot security gaps to fix","Get reliability and speed tips"],"pricing":"Free.","status":"ga","url":"https://azure.microsoft.com/en-us/products/advisor/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-backup","provider":"azure","name":"Azure Backup","category":"block-storage","tagline":"Automatic backups of your servers, databases and files in Azure.","description":"Azure Backup keeps safe copies of your data so you can get it back if something breaks or gets deleted. If your app runs on a virtual machine (a rented server in the cloud) or stores customer records in a database, you can set up regular backups without building your own system. It stores copies in a protected space called a vault, and it can also cover machines outside Azure. Pick it when you are already on Azure and want backups built in. You pay per protected item plus the storage the copies use.","useCases":["Back up virtual machines regularly","Protect database copies from loss","Back up files at scale"],"pricing":"Per protected instance plus storage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/backup/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-lighthouse","provider":"azure","name":"Azure Lighthouse","category":"management","tagline":"Let a trusted partner manage your Azure accounts without sharing logins.","description":"Azure Lighthouse lets one company manage Azure resources that belong to other companies, without needing separate logins for each. If you run an agency that looks after cloud setups for several clients, you can see and manage all of them from your own account. Customers choose exactly what access you get, can limit it to certain times, and can review a record of what you did. Pick it when you serve many customers and want tidy, controlled access instead of passing around passwords. It is free to use.","useCases":["Let partners manage client accounts","Run operations across many customers","Audit access for outside managers"],"pricing":"Free.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-lighthouse/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-managed-applications","provider":"azure","name":"Azure Managed Applications","category":"management","tagline":"Sell or share a ready-made Azure solution that runs in your customer's account.","description":"Azure Managed Applications lets you package a whole solution, then deliver it so it runs inside your customer's own Azure account. If you build a software product that needs several Azure pieces working together, you can offer it through Azure Marketplace (Microsoft's online store) and keep managing it for them. Companies can also use it to share approved setups internally. Pick it when customers want your product to live in their account, not yours. You set your own price, and Azure charges for the resources used.","useCases":["Sell packaged solutions through Marketplace","Deliver apps inside customer accounts","Offer an internal app catalog"],"pricing":"Publisher sets pricing; Azure charges for resources.","status":"ga","url":"https://azure.microsoft.com/en-us/products/managed-applications/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-migrate","provider":"azure","name":"Azure Migrate","category":"migration","tagline":"Plan and move your existing servers, databases and apps into Azure.","description":"Azure Migrate is the central place for moving what you already run into Azure. If your product currently lives on your own servers or another cloud, it can scan what you have, estimate what it will cost in Azure, and help you plan and carry out the move. It covers servers, databases and apps, with AI-assisted planning. Pick it when you want one guided toolset instead of piecing the move together yourself. The main tools are free, and you pay only for the Azure resources you end up using.","useCases":["Move servers into Azure","Plan database and app migrations","Estimate costs before moving"],"pricing":"No charge for core tooling; pay for resources.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-migrate/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-policy","provider":"azure","name":"Azure Policy","category":"management","tagline":"Set rules for your Azure account and catch anything that breaks them.","description":"Azure Policy lets you write rules for what can and cannot happen in your Azure account, then checks that everything follows them. If you want every resource labeled with a project name, or want to stop anyone from creating things in the wrong region, you set a rule once and it applies everywhere. It comes with ready-made rules and lets you create your own. Pick it when more than one person is building and you need guardrails to keep things tidy and compliant. It is free for Azure resources, though Arc (Microsoft's tool for outside machines) has charges.","useCases":["Require tags on all resources","Enforce compliance rules automatically","Block settings that break standards"],"pricing":"Free for Azure resources; Arc charges apply.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-policy/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-resource-manager","provider":"azure","name":"Azure Resource Manager","category":"management","tagline":"The control layer behind everything you create or change in Azure.","description":"Azure Resource Manager is the system underneath every action you take in Azure. When you click a button in the portal, run a command, or use a template, this is what receives the request and carries it out. If you want to set who can change what, or label resources by project, this is where that happens. You do not choose it instead of other tools, because they all go through it. That makes it useful to understand when you manage a growing account. It is free.","useCases":["Deploy resources through one control layer","Control who can access resources","Organize and tag cloud resources"],"pricing":"Free.","status":"ga","url":"https://azure.microsoft.com/en-us/get-started/azure-portal/resource-manager/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-resource-manager-templates","provider":"azure","name":"Azure Resource Manager templates","category":"devtools","tagline":"Describe your Azure setup in a file and rebuild it the same way every time.","description":"Azure Resource Manager templates let you write down your cloud setup in a file, so Azure builds it for you exactly as described. This approach is called infrastructure as code. If you need the same test and live environments for your app, you can run one template twice instead of clicking through menus and risking mistakes. You can write them in ARM templates or in Bicep, a simpler format for the same job. Pick them when you want repeatable setups you can save and share. They are free to use.","useCases":["Deploy the same setup repeatedly","Describe infrastructure as code","Build consistent environments for teams"],"pricing":"Free.","status":"ga","url":"https://azure.microsoft.com/en-us/products/arm-templates/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-site-recovery","provider":"azure","name":"Azure Site Recovery","category":"migration","tagline":"Keep your app running by copying it to a backup location in Azure.","description":"Azure Site Recovery makes a live copy of your servers in a second location, such as another Azure region, so your app can keep running if the first one fails. If an outage would cost you customers, you can switch over to the copy instead of rebuilding from scratch. This is called disaster recovery. Pick it when downtime hurts more than a simple data backup can fix, since a backup restores files but does not keep your app available. You pay a monthly fee for each protected machine.","useCases":["Keep apps running after a disaster","Fail over to a backup region","Replicate workloads to a second site"],"pricing":"Per protected instance per month.","status":"ga","url":"https://azure.microsoft.com/en-us/products/site-recovery/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-cloud-shell","provider":"azure","name":"Cloud Shell","category":"devtools","tagline":"A ready-to-use command line for Azure that runs in your web browser.","description":"Cloud Shell gives you a command line (a text window where you type instructions) inside your browser, already signed in to Azure. If you need to run a quick fix or follow a tutorial that says to type commands, you can open it and start without installing anything. It offers Bash or PowerShell, two common command languages, and comes with popular tools preinstalled. Microsoft keeps it updated for you. Pick it when you want to try something quickly from any computer. It is free, apart from a small storage account for your files.","useCases":["Run admin commands from a browser","Run scripts without installing anything","Manage cloud resources from any computer"],"pricing":"Free; pay for the small storage account.","status":"ga","url":"https://azure.microsoft.com/en-us/get-started/azure-portal/cloud-shell/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-cost-management","provider":"azure","name":"Microsoft Cost Management","category":"management","tagline":"See where your cloud money goes and set alerts before bills surprise you.","description":"Microsoft Cost Management shows what you are spending in the cloud and what is driving it. If your bill jumped last month and you do not know why, its cost analysis breaks it down by project or service. You can set budgets that alert you before you overspend, and get tips for cutting costs. It works across Azure and other clouds, and can split costs between teams. Pick it when you want clear spending answers without a spreadsheet. It is free for Azure usage.","useCases":["Set budget alerts for cloud spending","Split cloud bills between teams","Find ways to cut cloud costs"],"pricing":"Free for Azure usage.","status":"ga","url":"https://azure.microsoft.com/en-us/products/cost-management/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-network-watcher","provider":"azure","name":"Azure Network Watcher","category":"networking","tagline":"See what's happening on your Azure network and fix connection problems.","description":"Azure Network Watcher is a set of tools for seeing what goes on inside your Azure virtual networks, the private networks your cloud servers live in. If your app suddenly can't reach its database, you can use it to troubleshoot the connection, capture network traffic from a distance, or view a map of how everything is connected. It can also keep flow logs, which are records of which machines talked to which. Many of the tools are free, and some features are billed per use.","useCases":["Troubleshoot connection problems between servers","Record network traffic logs","Capture packets remotely to diagnose issues"],"pricing":"Free tools; some features billed per use.","status":"ga","url":"https://azure.microsoft.com/en-us/products/network-watcher/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-traffic-manager","provider":"azure","name":"Azure Traffic Manager","category":"dns","also":["load-balancing"],"tagline":"Send visitors to the nearest or healthiest copy of your app.","description":"Azure Traffic Manager directs visitors to the right copy of your app by answering DNS, the internet's address lookup, in a smart way. If you run your app in several regions and one goes down, it sends people to one that is still healthy. It can also route visitors to the location closest to them. Pick it when you want regional failover or location-based routing handled at the address-lookup level. You pay per DNS query and per endpoint it monitors.","useCases":["Send users to a healthy region","Fail over when a datacenter goes down","Route visitors by their location"],"pricing":"Per DNS query and monitored endpoint.","status":"ga","url":"https://azure.microsoft.com/en-us/products/traffic-manager/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-automanage","provider":"azure","name":"Azure Automanage","category":"management","tagline":"Keep your Azure servers patched, backed up and monitored automatically.","description":"Azure Automanage applies a ready-made set of settings to your virtual machines, which are servers you rent in the cloud, so they stay patched, backed up and monitored. If you run a handful of servers for your product and don't want to set up each of those chores by hand, you pick a profile and it does the setup for you. It suits hardening a single server or looking after a whole group of them. There is no extra charge for it, and you pay only for the resources you use.","useCases":["Apply standard patching and backup settings","Harden virtual machines automatically","Manage many servers with one profile"],"pricing":"No extra charge; pay for resources used.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-automanage/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-resource-mover","provider":"azure","name":"Azure Resource Mover","category":"migration","tagline":"Move your Azure resources from one region to another.","description":"Azure Resource Mover helps you relocate things you have built in Azure from one region, meaning a geographic location of Microsoft's data centers, to another. If most of your customers have shifted to Europe and you want your app closer to them, it plans the move, prepares it and carries it out. It analyzes dependencies, so pieces that rely on each other move in the right order. Pick it over moving things by hand when you have many connected resources. It has no extra charge.","useCases":["Move resources to another region","Plan moves with dependency checks","Expand services into new zones"],"pricing":"No extra charge.","status":"ga","url":"https://azure.microsoft.com/en-us/products/resource-mover/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-update-management-center","provider":"azure","name":"Update management center","category":"management","tagline":"One place to schedule and track software updates for your servers.","description":"Update management center, also called Azure Update Manager, is a single dashboard for installing software updates on your servers. If you run several servers and need to know which ones are missing security patches, you can see it at a glance and schedule updates for quiet hours. It covers servers in Azure and machines elsewhere that are connected through Azure Arc, a tool that lets Azure manage servers outside its own data centers. It is free for Azure virtual machines and billed per server for Arc machines.","useCases":["Track which machines need updates","Schedule software updates across servers","Enforce patching rules from one dashboard"],"pricing":"Free for Azure VMs; per server for Arc machines.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-update-management-center/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-content-delivery-network","provider":"azure","name":"Azure Content Delivery Network","category":"cdn","tagline":"Serve images and downloads quickly to visitors around the world.","description":"Azure Content Delivery Network keeps copies of your files on servers around the world so they load fast wherever your visitors are. If your site has large images, videos or software downloads, it hands each person the copy nearest to them instead of making them fetch it from one distant place. You pay per GB delivered. Microsoft is moving CDN features into Azure Front Door, so it is worth checking that service too before you commit.","useCases":["Serve images and scripts from nearby servers","Speed up file downloads worldwide","Deliver static website files quickly"],"pricing":"Pay per GB delivered.","status":"ga","url":"https://azure.microsoft.com/en-us/products/cdn/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-data-box","provider":"azure","name":"Azure Data Box","category":"migration","tagline":"Ship your big data to Azure on a physical device, not over the internet.","description":"Azure Data Box is a physical storage device that Microsoft ships to you, so you can fill it with data and send it back to be loaded into Azure. If you have huge piles of video, backups or old records, uploading over the internet could take a very long time, and this skips that wait. It also works for offline backups. Pick it when your internet connection is the bottleneck. You pay per order plus shipping.","useCases":["Ship large datasets on a physical device","Move huge data without slow internet","Load offline backups into cloud storage"],"pricing":"Per order plus shipping.","status":"ga","url":"https://azure.microsoft.com/en-us/products/databox/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-communication-services","provider":"azure","name":"Azure Communication Services","category":"messaging","also":["email","realtime"],"tagline":"Add chat, texting, calls, video and email to your app.","description":"Azure Communication Services gives you building blocks for adding chat, SMS text messages, voice calls, video and email to your own app. If you want customers to call a support agent from inside your product or get a text when their order ships, you plug these in instead of building them yourself. It is built on the same infrastructure as Microsoft Teams. Pick it when you want these features inside your product, not in a separate app. You pay per use, and rates vary by channel.","useCases":["Add voice and video calls to apps","Send SMS and email notifications","Build in-app chat messaging"],"pricing":"Pay per use; rates vary by channel.","status":"ga","url":"https://azure.microsoft.com/en-us/products/communication-services/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-application-gateway","provider":"azure","name":"Azure Application Gateway","category":"load-balancing","also":["waf"],"tagline":"Spread web traffic across your servers and block common attacks.","description":"Azure Application Gateway sits in front of your website or web app and decides which server should handle each visit. Because it understands web requests (layer 7, in network terms), it can send one page path to one group of servers and another path to a different group. It includes a web application firewall, a filter that blocks common attacks, and can do TLS offload, meaning it handles the encryption of connections so your servers don't have to. It works within one region. You pay hourly plus capacity units.","useCases":["Route web traffic to the right app","Block web attacks before they reach apps","Handle HTTPS encryption for backend servers"],"pricing":"Hourly plus capacity units.","status":"ga","url":"https://azure.microsoft.com/en-us/products/application-gateway","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-bastion","provider":"azure","name":"Azure Bastion","category":"networking","tagline":"Log in to your Azure servers from your browser, safely.","description":"Azure Bastion lets you log in to your cloud servers straight from your web browser, without giving those servers public internet addresses. If you need to fix something on a server, you connect through Bastion using RDP or SSH, the two standard remote-login methods, instead of opening the server to the whole internet. It can replace a jump box, a spare server people use as a stepping stone to reach others. It is a managed service, so Microsoft runs it for you. You pay hourly by tier plus outbound data.","useCases":["Log into servers without public addresses","Replace a jump box for admin access","Connect to virtual machines in a browser"],"pricing":"Hourly by SKU plus outbound data.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-bastion/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-ddos-protection","provider":"azure","name":"Azure DDoS Protection","category":"ddos","tagline":"Shield your public app from floods of fake traffic.","description":"Azure DDoS Protection defends your app against DDoS attacks, where attackers flood a site with fake traffic to knock it offline. If your product has a public website or API, it watches the traffic, tunes itself to what is normal for you, and works to stop the flood. Afterward you can read reports on what happened. It comes in network and IP protection tiers, so you can choose how much to cover. You pay a monthly fee per plan.","useCases":["Defend public sites from traffic floods","Get reports on attack attempts","Adjust protection to normal traffic"],"pricing":"Monthly fee per plan.","status":"ga","url":"https://azure.microsoft.com/en-us/products/ddos-protection/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-dns","provider":"azure","name":"Azure DNS","category":"dns","tagline":"Host your domain's address records on Microsoft's global network.","description":"Azure DNS hosts your domain's DNS records, the entries that tell the internet which server to visit when someone types your web address. If you own a domain for your product and want its records managed alongside the rest of your Azure setup, you add them here. It runs on Microsoft's global network with high availability and good performance. It also supports private zones, which are names that work only inside your own network. You pay per zone per month plus queries.","useCases":["Point your domain name to servers","Host private names inside your network","Manage DNS records for your domain"],"pricing":"Per zone per month plus queries.","status":"ga","url":"https://azure.microsoft.com/en-us/products/dns/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-firewall","provider":"azure","name":"Azure Firewall","category":"networking","also":["waf"],"tagline":"A managed security guard that filters traffic in and out of your network.","description":"Azure Firewall is a security checkpoint that Microsoft runs for you, deciding which network traffic may enter or leave your cloud setup. If your app talks to outside services and you want to limit which websites your servers can reach, you can set those rules here. It also uses threat intelligence, which means it can flag traffic from known bad sources, and it gives you analytics to see what is happening. Pick it over a do-it-yourself firewall on a server when you want something managed. You pay by the hour, based on the version you choose, plus the amount of data it processes.","useCases":["Control which sites servers can reach","Separate network parts from each other","Block known threats at the network edge"],"pricing":"Hourly by SKU plus data processed.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-firewall/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-route-server","provider":"azure","name":"Azure Route Server","category":"networking","tagline":"Lets your network appliances share routes with Azure automatically.","description":"Azure Route Server helps your network gear and your Azure network agree on where traffic should go, without you editing route tables by hand. If you run a third-party appliance, such as an SD-WAN device that links your offices to the cloud, it can announce its routes to Azure using BGP, a standard way for networks to swap directions. When something changes, Azure learns about it on its own. Choose it when your setup keeps changing and manual updates would be a chore. It is billed with an hourly fee.","useCases":["Let network appliances share routes","Connect SD-WAN devices to Azure","Update routing automatically between networks"],"pricing":"Hourly fee.","status":"ga","url":"https://azure.microsoft.com/en-us/products/route-server/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-web-application-firewall","provider":"azure","name":"Azure Web Application Firewall","category":"waf","also":["load-balancing"],"tagline":"Blocks common website attacks before they reach your app.","description":"Azure Web Application Firewall protects your website or web app from well-known attacks, such as SQL injection (sneaking database commands into a form) and cross-site scripting (planting harmful code in a page). If your app has login forms or a checkout, it can screen incoming requests and turn away the nasty ones. It includes rules based on OWASP, a respected list of top web risks, plus bot rules. It runs on Azure Front Door and Application Gateway, so you add it to those rather than buying it alone. Pricing is included with the gateway, plus rule fees.","useCases":["Block SQL injection and cross-site scripting","Filter bad bots from web apps","Apply standard web security rules"],"pricing":"Included with gateway pricing plus rule fees.","status":"ga","url":"https://azure.microsoft.com/en-us/products/web-application-firewall/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-front-door","provider":"azure","name":"Azure Front Door","category":"cdn","also":["load-balancing","waf"],"tagline":"Make your site fast worldwide, spread traffic out, and block attacks.","description":"Azure Front Door sits in front of your website or app and serves visitors from locations close to them, so pages load faster wherever they are. If your customers are in several countries, it can also send each visitor to the nearest healthy copy of your app, which is called load balancing. It includes a web application firewall that blocks common attacks, and it speeds up pages that change often. Pick it when you want caching, traffic routing and protection in one service. Pricing is a base fee plus charges per request and for data.","useCases":["Serve a global web app quickly","Balance traffic across regions","Cache content near users worldwide"],"pricing":"Base fee plus per request and data.","status":"ga","url":"https://azure.microsoft.com/en-us/products/frontdoor/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-virtual-network-manager","provider":"azure","name":"Azure Virtual Network Manager","category":"networking","tagline":"Set network and security rules once and apply them across many accounts.","description":"Azure Virtual Network Manager lets you control many Azure private networks from one place instead of setting each up by hand. If your company has grown to many subscriptions (separate billing and project containers in Azure), you can define how networks connect and which security rules always apply, then roll them out everywhere. It supports hub and spoke layouts, where one central network links to several smaller ones, and security admin rules that override local settings. Pick it when managing networks one at a time gets messy. It is billed per subscription hour.","useCases":["Apply network rules across many subscriptions","Set up hub and spoke networks","Enforce security rules at scale"],"pricing":"Per subscription hour.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-network-manager","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-nat-gateway","provider":"azure","name":"Azure NAT Gateway","category":"networking","tagline":"Give your private servers a safe, fixed address for going online.","description":"Azure NAT Gateway lets servers in a private part of your network reach the internet without letting the internet reach them. If your app calls outside services, such as a payment provider or an email API, the requests leave through this gateway. It uses static IPs, meaning the address never changes, so partners can add you to an allow list. It is built for low latency, so connections stay quick. Pick it over default outbound access when you need a predictable address. You pay an hourly fee plus a charge for the data it processes.","useCases":["Give private servers outbound internet access","Use fixed IP addresses for outgoing traffic","Let servers call third-party APIs"],"pricing":"Hourly plus data processed.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-nat-gateway","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-load-balancer","provider":"azure","name":"Azure Load Balancer","category":"load-balancing","tagline":"Spread incoming traffic across several servers so your app stays up.","description":"Azure Load Balancer shares incoming connections among several virtual machines (servers you rent in the cloud), so no single one gets overwhelmed. If your app suddenly gets busy, it sends each new visitor to a server that is ready, and skips any that are down. It works at layer 4, which means it handles raw TCP and UDP traffic, the basic ways computers send data, rather than reading web pages. It comes in Standard and Gateway versions. Pick it for fast, simple balancing. Pricing is per rule plus data processed.","useCases":["Spread traffic across virtual machines","Keep apps running if one server fails","Balance TCP and UDP connections"],"pricing":"Per rule and data processed.","status":"ga","url":"https://azure.microsoft.com/en-us/products/load-balancer/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-private-link","provider":"azure","name":"Azure Private Link","category":"networking","tagline":"Reach Azure services over a private path instead of the public internet.","description":"Azure Private Link gives you a private connection to Azure services and partner services, so your traffic stays on the Microsoft network and never crosses the open internet. If your app uses an Azure database, you can make that database reachable only from inside your own network. You do this with private endpoints, which act like a private door with an address in your network. You can also use it to offer your own service to others privately. Pick it when security rules say data must not travel publicly. You pay per endpoint hour plus data.","useCases":["Connect to databases without public internet","Keep service traffic on private network","Expose a service privately to partners"],"pricing":"Per endpoint hour plus data.","status":"ga","url":"https://azure.microsoft.com/en-us/products/private-link/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-firewall-manager","provider":"azure","name":"Azure Firewall Manager","category":"networking","tagline":"Manage the rules of many firewalls from a single place.","description":"Azure Firewall Manager is a control panel for managing several Azure Firewalls together. If you run firewalls in different regions or accounts, you can write a security policy once and apply it to all of them, instead of repeating the work. It also manages secured hubs, which are central network points with a firewall built in, and handles the routes that send traffic through them. Pick it when you have more than one firewall and want consistent rules. It does not replace the firewall itself. Pricing is per policy and deployment.","useCases":["Manage firewall rules from one place","Set policies across multiple firewalls","Organize secured network hubs"],"pricing":"Per policy and deployment.","status":"ga","url":"https://azure.microsoft.com/en-us/products/firewall-manager/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-virtual-network","provider":"azure","name":"Azure Virtual Network","category":"networking","tagline":"Your own private network inside Azure, where your services connect safely.","description":"Azure Virtual Network is the base layer for private networking in Azure, like building your own walled office network in the cloud. If you run a web app, a database and a few servers, you place them in one network so they can talk to each other while staying hidden from the public. You can split it into subnets (smaller sections), link it to other networks through peering, and connect to your office with gateways. Almost everything else in Azure networking builds on it. It costs nothing to use, though limits apply per subscription.","useCases":["Create private networks for your apps","Split networks into subnets","Connect your cloud to office networks"],"pricing":"Free of charge; limits per subscription.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-network/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-virtual-wan","provider":"azure","name":"Azure Virtual WAN","category":"networking","tagline":"Connect offices, remote users and Azure in one managed network.","description":"Azure Virtual WAN brings the links between your branch offices, remote workers and Azure into one managed service. If your business has several locations that each need to reach your cloud apps, it automates much of the setup and tunes the connections for you. It works with VPN, ExpressRoute (a private line into Azure) and SD-WAN partners, which are vendors that manage office internet links. Pick it over wiring each site by hand when you have many locations. Pricing is per hub and per connection unit, which measures how much capacity you use.","useCases":["Connect branch offices to cloud networks","Link VPNs and dedicated lines together","Automate networking between many sites"],"pricing":"Per hub and per connection unit.","status":"ga","url":"https://azure.microsoft.com/en-us/products/virtual-wan/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-vpn-gateway","provider":"azure","name":"Azure VPN Gateway","category":"networking","tagline":"Connect your office or laptop to Azure through an encrypted tunnel.","description":"Azure VPN Gateway creates an encrypted tunnel between your own network and Azure, so data travels privately across the public internet. If your team has an office network that needs to reach your cloud servers, you can link them with a site-to-site connection. If individual people need to log in from home, a point-to-point connection lets each person connect from their own device. Pick it when you want a secure link without paying for a dedicated private line. It is billed hourly, and the price depends on the gateway size you choose.","useCases":["Connect office networks to Azure privately","Let remote users connect securely","Encrypt traffic between sites and cloud"],"pricing":"Hourly by gateway SKU.","status":"ga","url":"https://azure.microsoft.com/en-us/products/vpn-gateway/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-enclave","provider":"azure","name":"Azure Enclave","category":"specialized","tagline":"Locked-down cloud setups for sensitive, heavily regulated work.","description":"Azure Enclave is a locked-down area of the cloud for work that comes with strict rules about who can see and handle the data. If you are building a product for a government agency or a regulated industry, it gives you built-in governance (rules and controls over who can do what) and protection for your workloads from the start. Pick it over a regular Azure setup when meeting compliance demands is the main concern and you would rather not piece together the guardrails yourself. Pricing is based on usage.","useCases":["Run regulated workloads in secured environments","Build secured setups for compliance-heavy projects","Protect sensitive apps with built-in governance"],"pricing":"Usage-based; see the pricing page.","status":"ga","url":"https://azure.microsoft.com/en-us/products/enclave","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-multicloud-interconnect","provider":"azure","name":"Azure Multicloud Interconnect","category":"networking","tagline":"A private link between Azure and AWS, so your data skips the internet.","description":"Azure Multicloud Interconnect is a private connection between Azure and other clouds, so your systems can talk to each other without going over the public internet. If your app runs partly on Azure and partly on AWS, it links Azure ExpressRoute (Azure's private network line) with AWS Interconnect at 1 Gbps (a fast connection speed, about a gigabit per second). It is a good fit when you want two clouds working together rather than moving everything to one. It is still in preview and free while the public preview lasts.","useCases":["Link Azure and AWS privately","Connect apps spanning two clouds","Test cross-cloud links during preview"],"pricing":"Free during public preview.","status":"preview","url":"https://azure.microsoft.com/en-us/products/multicloud-interconnect","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-cloud-hsm","provider":"azure","name":"Azure Cloud HSM","category":"secrets","tagline":"Your own locked box for encryption keys, inside your private network.","description":"Azure Cloud HSM gives you your own HSM (hardware security module, a tamper-resistant device that guards encryption keys) that nobody else shares, placed inside your virtual network (your private section of Azure). If your business must prove it alone controls its keys, or you are setting up the root of a PKI (the system that issues digital certificates), this is the fit. Choose it over Azure Key Vault when you need dedicated hardware that you own and control. You pay by the hour for each HSM.","useCases":["Hold your own encryption keys","Run cryptographic operations on dedicated hardware","Keep certificate authority root keys"],"pricing":"Hourly per HSM.","status":"ga","url":"https://azure.microsoft.com/en-us/products/cloud-hsm","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-dedicated-hsm","provider":"azure","name":"Azure Dedicated HSM","category":"secrets","tagline":"Older dedicated key-protection hardware, being retired.","description":"Azure Dedicated HSM is an older service that gave customers their own physical device (an HSM, or hardware security module) for guarding encryption keys. If your team already keeps keys on it, you can keep going for now, but it is only open to existing customers and there is no new onboarding. For anything new, Microsoft recommends Azure Cloud HSM or Managed HSM instead. Support continues until July 31, 2028.","useCases":["Keep existing hardware security modules running","Maintain current key storage deployments","Support existing HSM setups until 2028"],"pricing":"Existing customers only.","status":"deprecated","url":"https://learn.microsoft.com/en-us/azure/dedicated-hsm/overview","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-key-vault","provider":"azure","name":"Azure Key Vault","category":"secrets","also":["certificates"],"tagline":"A safe place to keep your app's passwords, keys and certificates.","description":"Azure Key Vault is a secure vault for the secrets your app needs, like passwords, API keys and encryption keys. If your app connects to a payment service or a database, you can store those credentials here instead of pasting them into your code. It comes in Standard, Premium and Managed HSM options (HSM means a tamper-resistant hardware device for keys), so you can pick the level of protection you need. Pricing is per operation, and HSM-backed options are priced separately.","useCases":["Store app passwords and secrets","Manage encryption keys and certificates","Safeguard passwords, keys and secrets"],"pricing":"Per operation; HSM-backed priced separately.","status":"ga","url":"https://azure.microsoft.com/en-us/products/key-vault/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-confidential-ledger","provider":"azure","name":"Azure confidential ledger","category":"specialized","tagline":"Tamper-proof record keeping for logs and compliance evidence.","description":"Azure confidential ledger is a place to store records that cannot be quietly changed after the fact. It runs inside trusted execution environments (specially protected areas of a computer chip) and gives you verifiable evidence that your data is intact. If your product needs audit logs or compliance records that customers or regulators can trust, this is where they could live. Pick it over an ordinary database when proving that nothing was altered matters most. You pay per ledger unit per hour.","useCases":["Keep tamper-proof audit logs","Store compliance records that can't be altered","Prove records were not changed"],"pricing":"Per ledger unit per hour.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-confidential-ledger/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-microsoft-azure-attestation","provider":"azure","name":"Microsoft Azure Attestation","category":"secrets","tagline":"Checks that a computer is trustworthy before it gets your secrets.","description":"Microsoft Azure Attestation is a checking service that confirms a machine is safe before you hand it anything sensitive. It verifies TPMs (security chips built into computers) and trusted execution environments (protected areas of a processor) before secrets are released. If you run confidential computing, where data stays protected even while it is being processed, you can use it for secure key release, so keys only go to machines that pass the check. You pay per attestation.","useCases":["Verify servers are trustworthy before use","Release secrets only to verified systems","Verify trusted hardware before sharing secrets"],"pricing":"Pay per attestation.","status":"ga","url":"https://azure.microsoft.com/en-us/products/azure-attestation/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-archive-storage","provider":"azure","name":"Archive Storage","category":"block-storage","tagline":"The cheapest place to park files you rarely need to open.","description":"Archive Storage is the lowest-cost tier of Azure Blob Storage, built for data you must keep but almost never read. If you need to hold compliance records or old backups for years, it has the lowest per-GB price in Blob Storage. The tradeoff is that getting files back takes hours, a step called rehydration, and it comes with its own fees. Pick it over regular storage only when you can afford to wait, since the savings come from not needing quick access.","useCases":["Keep rarely used data cheaply","Store long-term compliance archives","Restore archived files after hours-long wait"],"pricing":"Lowest per-GB price plus rehydration fees.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/blobs/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-managed-lustre","provider":"azure","name":"Azure Managed Lustre","category":"block-storage","also":["gpu"],"tagline":"A very fast shared drive for AI training and heavy number crunching.","description":"Azure Managed Lustre is a fully managed shared file system (a drive that many servers can read and write at once) built for high throughput. If you are training an AI model and need to feed it huge amounts of data quickly, or you need fast scratch space (temporary working storage) for HPC, meaning heavy scientific or technical calculations, it keeps up. It is built on Lustre, a file system popular in high-performance computing, and Azure runs it so you do not have to. You pay per TiB provisioned each month.","useCases":["Feed AI training data quickly","Provide fast scratch space for supercomputing","Run high-speed file storage without managing it"],"pricing":"Per TiB provisioned per month.","status":"ga","url":"https://azure.microsoft.com/en-us/products/managed-lustre/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-files","provider":"azure","name":"Azure Files","category":"block-storage","tagline":"Shared folders in the cloud that work like a network drive.","description":"Azure Files gives you shared folders in the cloud that behave like a network drive. They use SMB and NFS (two common ways computers share files) and work from Windows, Linux and macOS. If your team has an office file server you want to move to the cloud, or your app needs several servers to read the same files, it fits well. It is serverless, meaning there is no machine for you to look after. Pricing is per GiB, provisioned or used, depending on the tier.","useCases":["Share files across Windows, Linux and macOS","Move existing file shares to cloud","Provide shared storage for apps"],"pricing":"Per GiB provisioned or used, by tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/files/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-netapp-files","provider":"azure","name":"Azure NetApp Files","category":"block-storage","tagline":"Fast, dependable file storage for big business software and databases.","description":"Azure NetApp Files is high-performance file storage built on NetApp technology, with quick response times and built-in protection. If you run SAP (business software for finance and operations), a large database or heavy computing jobs, it gives you the speed those workloads expect. Pick it over Azure Files when performance is the priority and you are running demanding business software. You pay per GiB per hour, depending on the service level you choose.","useCases":["Run SAP with enterprise file storage","Host databases on low-latency file storage","Serve high-performance computing workloads with fast shares"],"pricing":"Per GiB per hour by service level.","status":"ga","url":"https://azure.microsoft.com/en-us/products/netapp/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-blob-storage","provider":"azure","name":"Azure Blob Storage","category":"object-storage","tagline":"Cloud storage for files, images and backups that grows with you.","description":"Azure Blob Storage is cloud storage for files of any kind, like photos, videos, backups and raw data. If your app lets people upload photos, this is where those images can live. It offers hot, cool, cold and archive tiers, so everyday files stay quick to reach while rarely used ones cost less to keep. It is the obvious starting point on Azure when you need to store files and do not need a shared drive. You pay per GB stored, by tier, plus operations.","useCases":["Store photos, videos and other files","Keep nightly backups at low cost","Build data lakes for analytics"],"pricing":"Per GB stored by tier plus operations.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/blobs/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-disk-storage","provider":"azure","name":"Azure Disk Storage","category":"block-storage","tagline":"Hard drives for your Azure servers, in fast and budget-friendly types.","description":"Azure Disk Storage gives your Azure servers (called virtual machines) their own hard drives, from the fastest Ultra disks to cheaper traditional ones. If you run a database for your app, or just need somewhere for a server's system and files to live, this is where that data sits. Pick it when you want a drive attached to a single server, rather than a shared folder or a pile of uploaded files. You pay per GiB (a unit of storage size) each month, and the price depends on the disk type.","useCases":["Attach fast disks to virtual machines","Power databases with high-speed disks","Choose SSD or HDD by need"],"pricing":"Per GiB per month by disk type.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/disks/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-elastic-san","provider":"azure","name":"Azure Elastic SAN","category":"block-storage","tagline":"One shared pool of fast storage for many apps and servers.","description":"Azure Elastic SAN is a big pool of storage that many of your servers and apps can share, instead of setting up a separate disk for each one. If you're moving from a company data center that already uses a SAN (a shared storage network), or running several databases that each need space, you can take capacity from one pool. It connects over iSCSI, a standard way for servers to talk to storage over a network. Pick it over single disks when you want to manage capacity in one place. You pay per TiB (a unit of storage size) you set aside.","useCases":["Move SAN-based workloads to Azure","Pool storage capacity across several apps","Run databases on shared block storage"],"pricing":"Per TiB provisioned.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/elastic-san/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-queue-storage","provider":"azure","name":"Queue Storage","category":"queues","tagline":"A simple waiting line for your app's background jobs.","description":"Queue Storage is a waiting line for messages that lives inside an Azure storage account. If your app lets people upload photos, you can drop a note in the queue for each one, and a separate worker picks them up later to resize them, so your users aren't left waiting. It holds large numbers of messages and is reached over HTTP, the same way web pages are. Pick it when you want a simple queue without setting up a heavier messaging system. You pay per GB stored plus per 10,000 operations.","useCases":["Queue tasks for background processing","Process work asynchronously between services","Pass large volumes of messages"],"pricing":"Per GB plus per 10,000 operations.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/queues/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-storage-accounts","provider":"azure","name":"Storage Accounts","category":"object-storage","tagline":"The account that holds your Azure files, queues and tables.","description":"A Storage Account is the top-level container that groups Azure's core storage services: Blob (files and images), Files (shared folders), Queue (message lines) and Table (simple data tables). If you're starting an app that needs to keep uploads, shared documents or background job messages, you create one account first and then use whichever of these you need inside it. Think of it as the building, with each service a room. Pick it as your starting point for general-purpose storage on Azure. Pricing is per service within the account.","useCases":["Group storage services under one account","Hold blobs, files, queues and tables","Set up general-purpose storage containers"],"pricing":"Per service within the account.","status":"ga","url":"https://azure.microsoft.com/en-us/products/category/storage/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-storage-explorer","provider":"azure","name":"Storage Explorer","category":"devtools","tagline":"A free desktop app to browse and manage your Azure storage.","description":"Storage Explorer is a free app for Windows, macOS and Linux that lets you click through your Azure storage like folders on your computer. If your app saves user uploads as blobs (files stored in the cloud), you can open the app to check they arrived, drag in a new file, or download one to take a look. Pick it when you want a quick visual view instead of writing code or typing commands. It manages your storage accounts and costs nothing to use.","useCases":["Browse files in storage accounts","Upload and download files from desktop","Manage storage from Windows, Mac or Linux"],"pricing":"Free.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage/storage-explorer/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-storage-discovery","provider":"azure","name":"Azure Storage Discovery","category":"management","tagline":"See how much storage you use, and what it's doing, across accounts.","description":"Azure Storage Discovery gives you one view of your storage across many accounts and subscriptions (the billing containers Azure uses to organize your resources). It shows capacity, activity and configuration side by side. If your storage bill keeps growing and you can't tell why, you can use it to spot which accounts are growing fastest and where you might cut back. Pick it over checking each account one by one once your storage is spread out. Pricing is usage-based.","useCases":["Track storage capacity trends over time","Spot cost savings across subscriptions","See activity and configuration across accounts"],"pricing":"Usage-based.","status":"ga","url":"https://azure.microsoft.com/en-us/products/storage-discovery","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-lab-services","provider":"azure","name":"Azure Lab Services","category":"end-user","tagline":"Ready-made virtual computer labs for classes, now being retired.","description":"Azure Lab Services lets teachers and trainers set up virtual computers in the cloud, so every student gets the same ready-to-use machine. If you ran a coding workshop, each student could log in to an identical setup without installing anything on their own laptop. It suits classrooms and training rather than everyday app hosting, and it is open only to existing customers. New customers cannot sign up since July 15, 2024, and the service retires on June 28, 2027.","useCases":["Keep existing classroom labs running","Run training labs already set up","Use existing virtual labs until 2027"],"pricing":"Existing customers only.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/lab-services/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-api-center","provider":"azure","name":"Azure API Center","category":"devtools","tagline":"A catalog of every API your team builds or uses, in one place.","description":"Azure API Center is a central list of all the APIs your organization builds or uses. An API is a way for one piece of software to talk to another. If your team has grown to several apps and nobody is sure which APIs already exist, you can register them here so people can find and reuse them. Pick it when scattered APIs are getting hard to track, since it also helps with governance (making sure everyone follows the same standards). It has Free and Standard plans.","useCases":["Catalog every API across the organization","Find existing APIs before building new ones","Govern API use across teams"],"pricing":"Free and Standard plans.","status":"ga","url":"https://learn.microsoft.com/en-us/azure/api-center/overview","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-kubernetes-fleet-manager","provider":"azure","name":"Azure Kubernetes Fleet Manager","category":"containers","tagline":"Update and manage many container setups on Azure from one place.","description":"Azure Kubernetes Fleet Manager lets you look after many AKS clusters as one group. AKS is Azure's managed Kubernetes, a system that runs your app's containers (packaged apps) across a set of servers. If you run separate clusters for different regions or customers, you can roll out updates, apply the same rules and decide which cluster runs which workload from one place. Pick it when managing clusters one by one gets tedious. It is free, with a charge for advanced features.","useCases":["Update several AKS clusters at once","Place workloads across multiple clusters","Apply governance across cluster groups"],"pricing":"Free; paid for advanced features.","status":"ga","url":"https://azure.microsoft.com/en-us/products/kubernetes-fleet-manager/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-compute-fleet","provider":"azure","name":"Azure Compute Fleet","category":"vms","tagline":"Launch a big mix of discounted and regular Azure servers at once.","description":"Azure Compute Fleet lets you launch a large group of servers (virtual machines) in one go, mixing Spot VMs, which are discounted spare capacity that Azure can take back, with standard ones. If you need to process a big batch of work, like converting thousands of videos overnight, it can spread servers across different sizes and zones (separate data center locations) to find capacity at a good price. Pick it over launching servers one at a time when cost and availability both matter. There is no extra charge, and you pay only for the VMs.","useCases":["Launch large batches of virtual machines","Mix Spot and standard machines for savings","Scale compute across zones and sizes"],"pricing":"No extra charge; pay for VMs.","status":"ga","url":"https://learn.microsoft.com/en-us/azure/azure-compute-fleet/overview","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-service-fabric","provider":"azure","name":"Azure Service Fabric","category":"containers","tagline":"Run an app as many small services that can keep their own data.","description":"Azure Service Fabric is a platform for building apps out of many small services (called microservices) that work together, including ones that keep their own data. Many Microsoft services run on it. If you're maintaining a large app already split into small pieces, where each piece needs to remember things like a shopping cart, it fits well. Pick it when you need services that hold their own state, or when you already run apps built on it. There is no extra charge for the platform, and you pay for the servers it runs on.","useCases":["Build stateful microservices","Run legacy platform applications","Run distributed microservice systems reliably"],"pricing":"No extra charge; pay for VMs.","status":"ga","url":"https://learn.microsoft.com/en-us/azure/service-fabric/service-fabric-overview","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-blueprints","provider":"azure","name":"Azure Blueprints","category":"management","tagline":"Governance templates, being retired on January 31, 2027.","description":"Azure Blueprints was a way to package your Azure setup rules and templates so new projects start with the same guardrails. If you manage several teams, you could assign one blueprint to each so every environment follows your standards. It is still labeled preview, an early-access stage. Deployment Stacks and template specs replace it, so use those for anything new. Azure Blueprints retires on January 31, 2027.","useCases":["Keep existing blueprint assignments working","Maintain governance templates already applied","Plan a move to Deployment Stacks"],"pricing":"Free.","status":"deprecated","url":"https://azure.microsoft.com/en-us/products/blueprints/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-artifact-signing","provider":"azure","name":"Artifact Signing","category":"certificates","tagline":"Digitally sign your apps and documents so people can trust them.","description":"Artifact Signing, formerly called Azure Trusted Signing, is a managed service that puts a digital signature on your software and documents, so people can tell they really came from you and were not changed along the way. If you ship a Windows app, you would use it to sign each release before people download it. It handles the certificates (the digital ID behind a signature), including their upkeep, so you do not have to manage them yourself. You pay a monthly fee that depends on the tier you choose.","useCases":["Sign Windows apps and installers","Sign documents with managed certificates","Skip manual certificate renewals"],"pricing":"Monthly fee by tier.","status":"ga","url":"https://azure.microsoft.com/en-us/products/artifact-signing","source":"https://azure.microsoft.com/en-us/products/"},{"id":"azure-playfab","provider":"azure","name":"Azure PlayFab","category":"specialized","tagline":"Ready-made backend for online games: players, matchmaking and multiplayer.","description":"Azure PlayFab is a ready-made backend for games that keep running after launch. It handles the behind-the-scenes work like saving player data, running in-game economies and supporting multiplayer. If you are building a mobile or PC game where people log in, earn items and play with friends, you can use it instead of building all of that yourself. It also includes matchmaking, which pairs players into the same game. There is a free tier to start, then you pay based on usage.","useCases":["Run backends for live games","Match players into multiplayer sessions","Manage player data and in-game economy"],"pricing":"Free tier; paid by usage.","status":"ga","url":"https://playfab.com/","source":"https://azure.microsoft.com/en-us/products/"},{"id":"gcp-gemini-enterprise-agent-platform","provider":"gcp","name":"Gemini Enterprise Agent Platform","category":"ai-platform","also":["agents","ai-models"],"tagline":"Google's main place to build, run and manage AI agents and models.","description":"Gemini Enterprise Agent Platform is Google Cloud's main AI platform, announced at Next 2026 as the evolution of Vertex AI. It brings together access to AI models, training your own, building AI agents (programs that take actions for you) and keeping them under control. If you want a support assistant that answers from your company's documents, or a custom model trained on your own data, you can do it all in one place. Pick it when you want everything under one roof on Google Cloud. Pricing is based on usage, with free credits for new accounts.","useCases":["Train custom machine learning models","Build agents that use company data","Fine-tune and evaluate Gemini models"],"pricing":"Usage-based per model, token, node hour or runtime. Free credits for new accounts.","status":"ga","url":"https://cloud.google.com/products/gemini-enterprise-agent-platform","source":"https://cloud.google.com/products"},{"id":"gcp-agent-platform-model-garden","provider":"gcp","name":"Model Garden","category":"ai-models","tagline":"A catalog of AI models from Google and others that you can try and use.","description":"Model Garden, formerly Vertex AI Model Garden, is a catalog of AI models from Google and others, including Gemini, Imagen, Veo, Claude, Llama and DeepSeek. If you are not sure which model suits your app, you can try several, then call your pick as a ready-to-use service or set it up on your own endpoint (a private address your app talks to). It is handy when you want Google, open and partner models under one bill. Prices are per token (a small chunk of text), per request or per hour of server time, depending on the model.","useCases":["Try several AI models in one place","Host open weights models yourself","Pay for partner AI models on one bill"],"pricing":"Per token, per request or per node hour depending on the model.","status":"ga","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/models","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/models"},{"id":"gcp-gemini-api-agent-platform","provider":"gcp","name":"Gemini models on Agent Platform","category":"ai-models","tagline":"Use Google's Gemini AI for text, images and video, with business controls.","description":"This is Google's Gemini family of AI models, plus its image, video and music generators, offered as a managed service (formerly Gemini on Vertex AI). If your app needs a chat assistant, document summaries or generated images, you send requests and get results back without running anything yourself. It adds business controls such as choosing where your data is stored and reserving capacity for steady traffic. Pick it over a basic AI connection when your company needs those controls. You pay per token (a small chunk of text) or per generated item.","useCases":["Power chat and summaries in production","Generate images and videos","Process batches of documents"],"pricing":"Per token or per generated item. Provisioned Throughput available for steady load.","status":"ga","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-studio","provider":"gcp","name":"Agent Studio","category":"ai-platform","tagline":"A workspace to test AI prompts, compare models and export working code.","description":"Agent Studio, formerly Vertex AI Studio, is a place in the Google Cloud console to experiment with AI before you build anything big. If you are writing the instructions (prompts) for a chatbot, you can try different versions, adjust settings and compare models side by side. When something works, you export the code and drop it into your app. Use it to find out what works before you commit to building. It has no separate charge, and you only pay for the model usage.","useCases":["Test and tune prompts","Compare AI models side by side","Build simple apps from prompts"],"pricing":"No separate charge. You pay for model usage.","status":"ga","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-runtime","provider":"gcp","name":"Agent Runtime","category":"agents","tagline":"Hosting for your AI agents that handles scaling, memory and long tasks.","description":"Agent Runtime, formerly Vertex AI Agent Engine, is a place to host AI agents, the programs that take actions on your behalf, so they stay online for real users. If you built a booking assistant with a toolkit like LangChain or LlamaIndex, you can deploy it here and let Google deal with growing traffic. It also keeps track of conversations and memory, and can run tasks for up to 7 days. It works with agents built on any framework or on custom code. You pay for processor and memory time used, with sessions and Memory Bank priced separately.","useCases":["Host production AI agents","Run agents that take up to 7 days","Run teams of cooperating agents"],"pricing":"Usage-based on vCPU and memory time. Sessions and Memory Bank priced separately.","status":"ga","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-development-kit","provider":"gcp","name":"Agent Development Kit (ADK)","category":"devtools","also":["agents"],"tagline":"Free toolkit for writing AI agents in code and testing them locally.","description":"Agent Development Kit, or ADK, is a free, open source toolkit (its code is public) for writing AI agents in Python, Java, Go or TypeScript. If you want several agents that work together, such as one that researches and one that writes, ADK gives you the building blocks and lets you test them on your own computer. It is tuned for Gemini but works with other models too. When you are ready, you can deploy to Agent Runtime, Cloud Run or GKE (Google's ways of hosting apps). The toolkit costs nothing, and you pay only for models and hosting.","useCases":["Build teams of cooperating agents","Create assistants that call tools","Test agents locally before deploying"],"pricing":"Free open source. You pay for models and hosting.","status":"ga","url":"https://google.github.io/adk-docs/","source":"https://cloud.google.com/products"},{"id":"gcp-agent-registry","provider":"gcp","name":"Agent Registry","category":"agents","also":["management"],"tagline":"A shared list of your company's AI agents and tools, so teams can reuse them.","description":"Agent Registry is a central catalog where your organization records its AI agents, tools, skills and MCP servers (connectors that let agents use other apps and data). If several teams are building agents, it stops them from rebuilding the same thing and shows you what already exists. Because it feeds Google Cloud's access rules (IAM) and Agent Gateway, you can set rules based on which registered agent is asking. Pick it once more than a few agents are running and you need to keep track of them. See the product page for current pricing.","useCases":["List all internal agents in one place","Share MCP servers across teams","Set access rules by agent identity"],"pricing":"See product page for current pricing.","status":"ga","url":"https://docs.cloud.google.com/agent-registry/overview","source":"https://docs.cloud.google.com/agent-registry/overview"},{"id":"gcp-agent-gateway","provider":"gcp","name":"Agent Gateway","category":"ai-gateway","also":["agents","networking"],"tagline":"A checkpoint that checks what your AI agents are allowed to do.","description":"Agent Gateway is a checkpoint that sits between your AI agents and the tools they call, including MCP servers (connectors to other apps and data). Before a request goes through, it checks Agent Registry, your access rules (IAM) and content security policies. If you worry that an agent could leak customer data or use a tool it should not, this is where you block that. It gives you one central place for sign-in rules rather than setting them tool by tool. It is in preview, meaning still being tested, and pricing is on the product page.","useCases":["Control what agents can call","Stop sensitive data from leaking out","Manage sign-in for MCP tools centrally"],"pricing":"See product page for current pricing.","status":"preview","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-identity","provider":"gcp","name":"Agent Identity","category":"identity","also":["agents"],"tagline":"Give each AI agent its own ID, so you control and track what it can do.","description":"Agent Identity gives every AI agent its own verifiable identity, a bit like an employee badge, that Google Cloud's access rules (IAM) can target. If you run an agent that reads your files and calls outside tools, you can allow only what it needs and see what it did afterward. It also manages the sign-in details agents need for third party tools. Pick it over sharing one account across many agents, which makes it hard to know who did what. It is in preview, meaning still being tested, and pricing is on the product page.","useCases":["Give each agent its own identity","Track what each agent does","Connect agents to third party tools securely"],"pricing":"See product page for current pricing.","status":"preview","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-search","provider":"gcp","name":"Agent Search","category":"ai-search","tagline":"Google-quality search for your website and data, and for your AI chatbots.","description":"Agent Search, formerly Vertex AI Search, lets you add search to your own websites and data, whether that is tidy tables or messy documents. If you run a help center and want visitors to type a question and find the right article, you can build that here. It also supports RAG (letting an AI answer using your own content) and grounding, which keeps a chatbot's answers tied to your data. Pick it when you want Google-style search without building it yourself. Pricing is per query plus storage for your indexed data, with a free trial.","useCases":["Search your website and intranet","Ground chat agents in your data","Build search for media or healthcare"],"pricing":"Per query plus data index storage. Free trial available.","status":"ga","url":"https://cloud.google.com/products/gemini-enterprise-agent-platform/agent-search","source":"https://cloud.google.com/products"},{"id":"gcp-agent-platform-vector-search","provider":"gcp","name":"Vector Search","category":"vector","also":["ai-search"],"tagline":"Find similar items fast, like related products or matching documents.","description":"Vector Search is a database that finds things that mean something similar, rather than matching exact words. If you are building a shop that suggests similar products, or a chatbot that needs to pull the right passages from your documents, you store each item as an embedding (a list of numbers that captures its meaning) and search by closeness. It is built on Google's ScaNN search technology, handles billions of embeddings, and can mix keyword and meaning-based search in one query. Pick it when you expect a very large collection and need fast answers.","useCases":["Find similar items by meaning","Power product recommendations","Retrieve context for AI answers"],"pricing":"Per node hour for serving plus index build and storage.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/vector-search/overview","source":"https://docs.cloud.google.com/vertex-ai/docs/vector-search/overview"},{"id":"gcp-rag-engine","provider":"gcp","name":"RAG Engine","category":"ai-search","tagline":"Let an AI answer questions using your own documents.","description":"RAG Engine is a ready-made pipeline that lets an AI model answer using your own documents instead of guessing. If you want a support bot that answers from your help articles, you hand it the files and it splits them into chunks, turns them into embeddings (numbers that capture meaning), and fetches the best matches when someone asks a question. Those matches go to Gemini, so answers stay grounded in your content. You choose where the data is stored, such as Spanner or Vector Search, two Google databases, and you pay for embeddings, storage and that choice.","useCases":["Answer questions from your documents","Power support chatbots","Build internal knowledge assistants"],"pricing":"Pay for embeddings, storage and the vector backend you pick.","status":"ga","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-platform-pipelines","provider":"gcp","name":"Agent Platform Pipelines","category":"ai-platform","also":["workflows"],"tagline":"Run your AI training steps automatically, the same way every time.","description":"Agent Platform Pipelines chains the steps of an AI project, like preparing data, training and testing, into a routine that runs the same way each time. If you want a model retrained every week on fresh data, you set it up once and let it run on a schedule, with each step running in a container (a self-contained package of code). It keeps a record of which data and files produced which result, so you can reproduce past work. It is serverless, meaning there are no machines for you to manage. You pay per run plus the computing each step uses.","useCases":["Schedule regular model retraining","Track experiments and outputs","Check models before releasing them"],"pricing":"Per pipeline run plus the compute each step uses.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/pipelines/introduction","source":"https://docs.cloud.google.com/vertex-ai/docs/pipelines/introduction"},{"id":"gcp-agent-platform-workbench","provider":"gcp","name":"Agent Platform Workbench","category":"ai-platform","also":["devtools"],"tagline":"Notebooks in the cloud for exploring data and trying out AI models.","description":"Agent Platform Workbench gives you Jupyter notebooks, which are pages where you write and run code in small steps and see results right away, hosted for you in Google Cloud. If you are looking through sales data in BigQuery, Google's data warehouse, you can open a notebook and start without setting up your own machine. You can add GPUs (chips that speed up AI work) when a model needs them, and run notebooks on a schedule. Choose it when you want a managed notebook already connected to your Google data. You pay hourly for the instance's computing and disk.","useCases":["Explore data in notebooks","Prototype models on GPUs","Schedule notebook runs"],"pricing":"Compute and disk of the instance, billed per hour.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/workbench/introduction","source":"https://docs.cloud.google.com/vertex-ai/docs/workbench/introduction"},{"id":"gcp-colab-enterprise","provider":"gcp","name":"Colab Enterprise","category":"ai-platform","also":["devtools"],"tagline":"Shared notebooks for working with data, with company-grade security.","description":"Colab Enterprise is the Colab notebook, a page where you write and run code step by step, brought into Google Cloud with business security controls. If your team analyzes data in BigQuery, Google's data warehouse, you can work on the same notebook together and keep access limited to the right people through IAM (Google Cloud's permission system). It also offers Gemini assistance and works with BigQuery Studio. Pick it over a regular notebook when your data is sensitive and your company needs control over who sees what. The notebook screen is free, and you pay for the machine that runs your code.","useCases":["Share notebooks with your team","Analyze BigQuery data in notebooks","Build data apps in notebooks"],"pricing":"Runtime charges by machine type. Notebook UI is free.","status":"ga","url":"https://cloud.google.com/colab-enterprise","source":"https://cloud.google.com/products"},{"id":"gcp-agent-platform-feature-store","provider":"gcp","name":"Feature Store","category":"ai-platform","tagline":"Keep the data your AI models learn from ready to use and consistent.","description":"Feature Store is a shared shelf of features, the pieces of information an AI model uses to make a prediction, such as how many orders a customer placed this month. If you are building fraud checks or product recommendations, you build these features in BigQuery, Google's data warehouse, and serve them quickly when your app needs a decision. The same values are used for training and for live predictions, so the two do not drift apart. You pay for the servers that deliver features live plus BigQuery storage.","useCases":["Serve features for recommendations","Score fraud with live data","Keep training and serving data consistent"],"pricing":"Pay for online serving nodes and BigQuery storage.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/overview","source":"https://docs.cloud.google.com/vertex-ai/docs/featurestore/latest/overview"},{"id":"gcp-agent-platform-model-registry","provider":"gcp","name":"Model Registry","category":"ai-platform","tagline":"A home for your AI models, with versions, so you can ship the right one.","description":"Model Registry is a catalog where you keep each version of the AI models you train. If you have retrained a model five times and need to know which one is live, you can track versions, mark one as the current choice with an alias (a nickname), and move forward when a new one is better. You can send a registered model straight to a live endpoint, which is an address your app calls for answers, or run it over a big batch of data. There is no separate charge for it beyond storage and serving.","useCases":["Version and track ML models","Promote models to production","Run batch predictions from models"],"pricing":"No separate charge beyond storage and serving.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/model-registry/introduction","source":"https://docs.cloud.google.com/vertex-ai/docs/model-registry/introduction"},{"id":"gcp-agent-platform-managed-training","provider":"gcp","name":"Managed Training","category":"ai-platform","also":["gpu"],"tagline":"Rent the computing you need to train or fine-tune your own AI models.","description":"Managed Training lets you train AI models on Google's hardware without owning any. If you want to fine-tune an open model (adjusting an existing model so it fits your task) on your own data, you submit a job and it runs on GPUs or TPUs, which are chips built for AI work. It handles one-off jobs and large reserved clusters for big distributed training, and it can try many settings automatically to find the best one. You pay for the machine and chip hours the job uses.","useCases":["Fine-tune open models","Train large models across GPUs","Search for best training settings"],"pricing":"Per machine and accelerator hour used by the job.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/training/overview","source":"https://docs.cloud.google.com/vertex-ai/docs/training/overview"},{"id":"gcp-agent-platform-inference","provider":"gcp","name":"Agent Platform Inference","category":"ai-platform","tagline":"Turn a trained AI model into a live service your app can call.","description":"Agent Platform Inference hosts a trained model so your app can ask it for answers. If you built a model that scores incoming leads, you put it on an endpoint, an address your app calls in real time, and it adds or removes capacity as traffic changes. For large piles of data you can run a batch job instead. It supports GPUs, TPUs and ready-made serving containers, so it also works for open weights models, whose files are publicly released. Online endpoints cost per hour of server time, and batch jobs are billed on the computing used.","useCases":["Host real-time model APIs","Score large batches of data","Serve open weights models"],"pricing":"Per node hour for online endpoints. Batch billed on compute used.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/predictions/overview","source":"https://docs.cloud.google.com/vertex-ai/docs/predictions/overview"},{"id":"gcp-agent-platform-automl","provider":"gcp","name":"Agent Platform AutoML","category":"ai-platform","tagline":"Train an AI model on your own labeled examples without writing much code.","description":"Agent Platform AutoML builds an AI model for you from examples you have labeled. If you have product photos tagged by category, you upload them and it trains a model that sorts new photos, with no deep machine learning knowledge needed. It handles image, text, video and tabular (spreadsheet-style) data. Pick it when your team wants a custom model but has no AI specialists, rather than building one from scratch. You pay per hour of training and per hour a model is serving predictions.","useCases":["Classify images without writing code","Predict outcomes from tables","Label videos from your examples"],"pricing":"Per node hour of training and per prediction hour.","status":"ga","url":"https://docs.cloud.google.com/vertex-ai/docs/beginner/beginners-guide","source":"https://docs.cloud.google.com/vertex-ai/docs/beginner/beginners-guide"},{"id":"gcp-agent-platform-evals","provider":"gcp","name":"Gen AI Evaluation","category":"ai-platform","also":["ai-gateway"],"tagline":"Test how good your AI's answers are before your users see them.","description":"Gen AI Evaluation scores the answers your AI models and agents give. If you are changing a prompt (the instructions you give the AI) and want to be sure it did not make things worse, you run the same questions through the old and new versions and compare scores. It can grade with metrics, rubrics (checklists of what a good answer looks like) and LLM judges, which are AI models that grade other AI's work. That makes it useful for comparing models and versions before release. You pay based on the judge model's token usage, the unit AI providers use to count text.","useCases":["Run regression tests on prompts","Compare models before release","Score agent answer quality"],"pricing":"Usage-based on judge model tokens.","status":"ga","url":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview","source":"https://docs.cloud.google.com/gemini-enterprise-agent-platform/overview"},{"id":"gcp-agent-platform-vision","provider":"gcp","name":"Agent Platform Vision","category":"ai-apps","also":["video"],"tagline":"Build apps that watch camera video and count, detect or flag what happens.","description":"Agent Platform Vision helps you build apps that understand live video from cameras. If you run a store and want to know how busy each aisle gets, you connect camera streams and use ready-made or custom computer vision models (AI that interprets images) to count people. It takes in, analyzes and stores the video in one place. It also fits safety monitoring and occupancy counting. You pay per hour for each stream and for each model you use.","useCases":["Count people in a space","Monitor safety on camera feeds","Analyze retail store traffic"],"pricing":"Per stream hour and per model used.","status":"ga","url":"https://cloud.google.com/vertex-ai-vision","source":"https://cloud.google.com/products"},{"id":"gcp-gemini-enterprise","provider":"gcp","name":"Gemini Enterprise","category":"ai-apps","also":["agents"],"tagline":"A workplace AI assistant that can search and act on your company's data.","description":"Gemini Enterprise is an AI assistant for your whole team that can search your company's information and run small helper programs, called agents, to do tasks for you. If your files are spread across Google Workspace and Microsoft 365, employees can ask one place for answers instead of hunting through each app. People who do not code can also build their own agents. Pick it when you want AI for everyone on staff, not just a tool for developers. It is priced per person per month, and you contact sales for editions.","useCases":["Give staff an AI assistant","Build agents without writing code","Search across company apps"],"pricing":"Per user per month. Contact sales for editions.","status":"ga","url":"https://cloud.google.com/gemini-enterprise","source":"https://cloud.google.com/products"},{"id":"gcp-gemini-enterprise-cx","provider":"gcp","name":"Gemini Enterprise for Customer Experience","category":"ai-apps","also":["end-user"],"tagline":"AI shopping and customer service helpers for stores and service teams.","description":"Gemini Enterprise for Customer Experience is a bundle of AI tools that help online stores and service teams talk with customers. If you run an online shop, it can power a shopping assistant that chats with visitors and a product search that understands what they mean. It also covers customer service chat and phone agents. It builds on Dialogflow CX, Google's chatbot builder, and Vertex AI Search for Commerce, its store search product, so you get both in one package. Pricing is through sales.","useCases":["Run an AI shopping assistant","Automate customer service conversations","Power product search for shoppers"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/gemini-enterprise-cx","source":"https://cloud.google.com/products"},{"id":"gcp-cx-agent-studio","provider":"gcp","name":"Customer Experience Agent Studio","category":"ai-apps","also":["end-user"],"tagline":"Build chat and phone assistants that handle customer questions.","description":"Customer Experience Agent Studio is a tool for building virtual assistants that talk with customers by chat or voice. If you want a phone bot that tells people where their order is, you can lay out the conversation as a visual flow and let AI handle the open-ended parts. It also connects to phone systems. It is the current home of Dialogflow CX style building, so choose it over older Dialogflow editions for new, more complex projects. You pay per request, or per minute for audio.","useCases":["Build phone support bots","Create chat assistants for customers","Automate order status questions"],"pricing":"Per request or per minute for audio. Contact sales for volume.","status":"ga","url":"https://cloud.google.com/gemini-enterprise-cx/cx-agent-studio","source":"https://cloud.google.com/products"},{"id":"gcp-dialogflow-es","provider":"gcp","name":"Dialogflow ES","category":"ai-apps","tagline":"The original Google chatbot builder, best for simple bots.","description":"Dialogflow ES is the original edition of Google's tool for building chatbots and voice assistants. It works by matching what a person says to an intent, meaning the thing they want, such as asking for opening hours. If you need a basic FAQ bot for your website or a simple phone menu, it does the job and comes with a free trial. For anything larger or more complex, Google points new projects to CX Agent Studio instead. After the trial, you pay per text or audio request.","useCases":["Build basic FAQ chatbots","Run simple voice phone menus","Add voice actions to smart speakers"],"pricing":"Free trial. Per text or audio request after.","status":"ga","url":"https://cloud.google.com/dialogflow","source":"https://cloud.google.com/products"},{"id":"gcp-agent-assist","provider":"gcp","name":"Agent Assist","category":"ai-apps","also":["end-user"],"tagline":"Live answer suggestions and call summaries for your support staff.","description":"Agent Assist is an AI helper that sits beside your human customer support staff during a call or chat. It listens to the live conversation and shows suggested answers, help articles and summaries, so the person does not have to go searching. If your team handles tricky questions, it helps newer staff find the right answer quickly and cuts the time spent writing call notes. Pick it when you want to keep humans in the conversation instead of replacing them with a bot. Pricing is per conversation.","useCases":["Suggest live answers to support staff","Write call summaries automatically","Coach human agents during calls"],"pricing":"Per conversation. Contact sales for volume.","status":"ga","url":"https://cloud.google.com/agent-assist","source":"https://cloud.google.com/products"},{"id":"gcp-customer-experience-insights","provider":"gcp","name":"Customer Experience Insights","category":"ai-apps","also":["analytics"],"tagline":"Find patterns, moods and problems in your customer calls and chats.","description":"Customer Experience Insights turns your customer calls and chats into something you can study. It writes out what was said, tags the topics and sentiment (whether the customer sounded happy or upset), and highlights trends and quality gaps. If you have hundreds of support conversations and no time to read them, it can show you what customers keep asking about. It was formerly called Conversational Insights. Choose it when you want to learn from conversations you already have. You pay per conversation analyzed.","useCases":["Check support call quality","Find trending customer topics","Track customer mood over time"],"pricing":"Per conversation analyzed.","status":"ga","url":"https://cloud.google.com/contact-center/insights/docs","source":"https://cloud.google.com/products"},{"id":"gcp-contact-center-as-a-service","provider":"gcp","name":"Contact Center as a Service","category":"end-user","also":["ai-apps"],"tagline":"A cloud call and chat support center with AI built in.","description":"Contact Center as a Service is a ready-made customer support center that runs in the cloud. It sends calls and chats to the right person, gives your staff one screen to work from, and adds AI help along the way. If you are launching a product and need phone, chat and other channels in one place, you do not have to stitch separate tools together. It was formerly called Contact Center AI Platform, so older articles may use that name. Pricing is per seat, meaning per support person, through sales.","useCases":["Run support across phone and chat","Route calls and chats to agents","Give agents one shared workspace"],"pricing":"Per seat. Contact sales.","status":"ga","url":"https://cloud.google.com/solutions/contact-center-as-a-service-platform","source":"https://cloud.google.com/products"},{"id":"gcp-conversational-analytics-api","provider":"gcp","name":"Conversational Analytics API","category":"ai-apps","also":["warehouse"],"tagline":"Let your app answer plain-English questions about your business data.","description":"Conversational Analytics API lets your app or AI agent ask questions about data in everyday English and get back answers, charts and the SQL (database query) behind them. It works with BigQuery, Google's data warehouse, and Looker, its reporting tool. If you want a chat box in your product where teammates or customers can ask what sales looked like last week, this is the building block. It also powers the data agents inside BigQuery. Billing depends on how much you use it.","useCases":["Let users chat with their data","Add an analytics assistant to apps","Build agents that answer data questions"],"pricing":"Usage-based. See pricing page.","status":"ga","url":"https://docs.cloud.google.com/gemini/docs/conversational-analytics-api/overview","source":"https://docs.cloud.google.com/gemini/docs/conversational-analytics-api/overview"},{"id":"gcp-gemini-code-assist","provider":"gcp","name":"Gemini Code Assist","category":"devtools","also":["ai-apps"],"tagline":"An AI coding helper that works inside your editor and reviews your code.","description":"Gemini Code Assist is an AI helper that suggests code, answers questions and can carry out tasks with several steps while you build. It runs inside editors like VS Code and JetBrains and in Cloud Shell, Google's browser-based terminal. It also reviews pull requests, which are proposed code changes, on GitHub. If you build with AI and want a second set of eyes on each change, it can explain unfamiliar code and point out issues. There is a free tier for individuals. Standard and Enterprise plans are per user per month.","useCases":["Generate code inside your editor","Review pull requests automatically","Explain unfamiliar code clearly"],"pricing":"Free individual tier. Standard and Enterprise are per user per month.","status":"ga","url":"https://cloud.google.com/products/gemini/code-assist","source":"https://cloud.google.com/products"},{"id":"gcp-gemini-cloud-assist","provider":"gcp","name":"Gemini Cloud Assist","category":"management","also":["ai-apps","observability"],"tagline":"An AI helper in Google Cloud that answers questions and investigates issues.","description":"Gemini Cloud Assist is a chat helper built into the Google Cloud console, the website where you manage your cloud setup. You can ask it about your resources, have it investigate an incident, or get a draft configuration. If your app goes down and you are not sure why, you can describe the problem in plain words instead of reading through logs. It can also answer cost questions and help you plan your architecture. It is included with Gemini for Google Cloud, and the pricing page has details.","useCases":["Investigate cloud outages and incidents","Answer questions about cloud costs","Draft cloud setup configurations"],"pricing":"Included with Gemini for Google Cloud. See pricing page.","status":"ga","url":"https://cloud.google.com/products/gemini/cloud-assist","source":"https://cloud.google.com/products"},{"id":"gcp-google-antigravity","provider":"gcp","name":"Google Antigravity","category":"devtools","also":["agents"],"tagline":"A desktop app where several AI agents plan and write code for you.","description":"Google Antigravity is a desktop code editor and command line tool (a way to run typed commands) where several AI agents plan and carry out coding work. If you want to build an app by describing what you need, you can have different agents work on different parts at the same time. It is also where Google points people moving projects over from Firebase Studio. Google lists it with the Gemini Enterprise Agent Platform. There is a free tier, and paid use comes through Google AI plans or Agent Platform.","useCases":["Build apps with coding agents","Run several coding agents at once","Move Firebase Studio projects over"],"pricing":"Free tier. Paid via Google AI plans or Agent Platform.","status":"ga","url":"https://antigravity.google/","source":"https://cloud.google.com/products"},{"id":"gcp-google-cloud-mcp-servers","provider":"gcp","name":"Google Cloud MCP servers","category":"agents","also":["devtools"],"tagline":"Let AI agents use Google Cloud services like BigQuery and Cloud Run.","description":"Google Cloud MCP servers are ready-made connection points that let AI agents use Google Cloud services. MCP, short for Model Context Protocol, is a common standard for plugging AI tools into other software. If you want your coding assistant to look up data in BigQuery or manage something on Cloud Run, which runs your app online, these servers give it a way in. Google hosts them for services including BigQuery, GKE and Cloud Run. Access is controlled with IAM, Google Cloud's permission system. The servers are free, and the services behind them bill as usual.","useCases":["Let AI agents use BigQuery","Let agents run cloud operations","Connect coding tools to cloud services"],"pricing":"Server access free. Underlying services bill as usual.","status":"ga","url":"https://docs.cloud.google.com/mcp/overview","source":"https://docs.cloud.google.com/mcp/overview"},{"id":"gcp-model-armor","provider":"gcp","name":"Model Armor","category":"ai-gateway","also":["waf"],"tagline":"Check what goes into and comes out of your AI for attacks and leaks.","description":"Model Armor is a safety filter that checks what people send to your AI model and what it sends back. If you run a chatbot, it can catch prompt injection (tricking the AI into ignoring your rules), jailbreaks (attempts to bypass its limits), private data like names or card numbers, and harmful content before they reach a user. It works with any model, not only Google's Gemini, so the protection stays if you switch models. You pay per token (a small chunk of text) screened, with a free allotment on some tiers.","useCases":["Block prompt injection attacks","Filter personal data from responses","Screen unsafe content before it ships"],"pricing":"Per token screened. Free allotment on some tiers.","status":"ga","url":"https://cloud.google.com/security/products/model-armor","source":"https://cloud.google.com/products"},{"id":"gcp-speech-to-text","provider":"gcp","name":"Speech-to-Text","category":"ai-apps","tagline":"Turn spoken audio into written text in more than 125 languages.","description":"Speech-to-Text listens to audio and writes out what was said, either from a saved file or live as someone talks. If you run a call center tool or a video app, you can use it to create transcripts and captions automatically. It can tell speakers apart (called diarization) and learn your own special words, like product names, so they come out spelled right. It is built on Google's Chirp speech models. You pay per minute of audio, and there is a free monthly allowance.","useCases":["Transcribe customer call recordings","Add captions to videos","Accept spoken voice commands"],"pricing":"Per minute of audio. Free monthly allowance.","status":"ga","url":"https://cloud.google.com/speech-to-text","source":"https://cloud.google.com/products"},{"id":"gcp-text-to-speech","provider":"gcp","name":"Text-to-Speech","category":"ai-apps","tagline":"Turn written text into natural-sounding voices in many languages.","description":"Text-to-Speech reads your text aloud in a realistic synthetic voice. If you are building a voice assistant, an audiobook, or the automated phone menu people hear when they call a company, you can generate the speech instead of recording a person each time. It offers neural and Gemini-based voices, lets you control pacing and pronunciation with SSML (a simple markup for tweaking how words are spoken), and supports custom voice options. Pricing is per million characters, with a free monthly allowance.","useCases":["Generate voices for assistants","Narrate audiobooks in many languages","Create spoken phone menu prompts"],"pricing":"Per million characters. Free monthly allowance.","status":"ga","url":"https://cloud.google.com/text-to-speech","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-translation","provider":"gcp","name":"Cloud Translation","category":"ai-apps","tagline":"Translate text, documents and big batches of content by API.","description":"Cloud Translation converts text from one language to another through an API (a way for your app to ask Google's service for a result). If your app has chat between people who speak different languages, or you need to localize your website or help articles, you send the text and get the translation back. Basic and Advanced editions are available. Advanced adds glossaries (your own approved word choices), adaptive translation (which learns from your example sentences) and Gemini-based models. Pricing is per character, with a free monthly allowance.","useCases":["Translate website and app content","Translate chat messages live","Translate whole documents"],"pricing":"Per character. Free monthly allowance.","status":"ga","url":"https://cloud.google.com/translate","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-vision-api","provider":"gcp","name":"Vision API","category":"ai-apps","tagline":"Find text, objects, labels and unsafe content in images by API.","description":"Vision API looks at an image and tells you what is in it. If your app lets people upload photos, you can use it to read text in the picture (known as OCR), tag what it shows, spot landmarks and objects, or flag adult or violent content before it goes public. It uses ready-made models, so you do not need to train anything yourself. You pay per 1,000 images for each feature you use, and the first 1,000 each month are free.","useCases":["Pull text out of images","Flag unsafe pictures for moderation","Tag photos with labels automatically"],"pricing":"Per 1,000 images per feature. First 1,000 free monthly.","status":"ga","url":"https://cloud.google.com/vision","source":"https://cloud.google.com/products"},{"id":"gcp-video-intelligence-api","provider":"gcp","name":"Video Intelligence API","category":"ai-apps","also":["video"],"tagline":"Detect scenes, objects, text and explicit content in video.","description":"Video Intelligence API watches a video for you and describes what happens in it. If your app hosts user videos, you can use it to make them searchable by what appears on screen, tag them automatically, or flag explicit content for review. It works on stored files or streamed video and uses ready-made models, so there is nothing to train. Pricing is per minute of video for each feature you turn on.","useCases":["Make videos searchable by content","Flag unsafe video for moderation","Tag video files with metadata"],"pricing":"Per minute of video per feature.","status":"ga","url":"https://cloud.google.com/video-intelligence","source":"https://cloud.google.com/products"},{"id":"gcp-natural-language-api","provider":"gcp","name":"Natural Language API","category":"ai-apps","tagline":"Read text for mood, people, places and topics by API.","description":"Natural Language API reads text and tells you what it is about and how it feels. If you collect customer reviews or support messages, you can score each one as positive or negative (sentiment), pull out the names of people, companies and places (entities), and sort the text into topics. It uses ready-made models, so you can start without building your own. Pricing is per 1,000 characters, with a free monthly allowance.","useCases":["Track sentiment in customer text","Pull names and places from text","Sort text into content categories"],"pricing":"Per 1,000 characters units. Free monthly allowance.","status":"ga","url":"https://cloud.google.com/natural-language","source":"https://cloud.google.com/products"},{"id":"gcp-document-ai","provider":"gcp","name":"Document AI","category":"ai-apps","tagline":"Pull clean data out of invoices, forms and contracts.","description":"Document AI reads documents and turns them into organized data you can use, like invoice totals, names and dates. If your product handles uploaded invoices, ID documents for identity checks, or contracts, it saves you from typing fields in by hand. It combines OCR (reading text from scans and photos) with Gemini-based extraction, and offers both ready-made and custom processors, which are the document readers you set up for each type. Pricing is per page, depending on the processor type.","useCases":["Extract data from invoices automatically","Read identity and compliance documents","Pull key fields from contracts"],"pricing":"Per page, by processor type.","status":"ga","url":"https://cloud.google.com/document-ai","source":"https://cloud.google.com/products"},{"id":"gcp-document-ai-warehouse","provider":"gcp","name":"Document AI Warehouse","category":"ai-apps","tagline":"Document storage and search with AI tagging (discontinued).","description":"Document AI Warehouse was a place to store, tag and search large collections of documents, such as contracts. Teams used it to keep older document repositories organized with AI help. If you are starting something new, look at Agent Search for finding documents and Document AI for pulling data out of them. Document AI Warehouse is closed to new use and being retired, and only existing customers can still use it.","useCases":["Manage existing document repositories","Search legacy contract archives","Plan a move to Document AI"],"pricing":"Existing customers only.","status":"closed","url":"https://cloud.google.com/document-warehouse","source":"https://cloud.google.com/products"},{"id":"gcp-enterprise-knowledge-graph","provider":"gcp","name":"Enterprise Knowledge Graph","category":"ai-apps","tagline":"Match records that mean the same thing across your datasets.","description":"Enterprise Knowledge Graph figures out when different records describe the same real-world thing, such as two company names spelled differently. If you have customer or supplier lists spread across several BigQuery tables (Google's data warehouse), it can match them against each other or against Google's own knowledge graph, a large collection of known people, places and organizations. That makes it handy for cleaning messy master data into one trusted list. Pricing is per record processed.","useCases":["Match duplicate records across datasets","Clean up master customer data","Link entities from different datasets"],"pricing":"Per record processed.","status":"ga","url":"https://cloud.google.com/enterprise-knowledge-graph","source":"https://cloud.google.com/products"},{"id":"gcp-ai-commerce-search","provider":"gcp","name":"AI Commerce Search","category":"ai-search","also":["ai-apps"],"tagline":"Smarter search and recommendations for your online store.","description":"AI Commerce Search powers the search bar, browse pages and \"you might also like\" suggestions in an online store. It learns from your product catalog and from what shoppers click and buy, so results get better for your own customers. If you run an ecommerce shop and want results ranked to drive sales, it is a better fit than a general-purpose search tool. It was previously called Vertex AI Search for commerce. Pricing is through contacting sales or per request.","useCases":["Power ecommerce site search","Recommend products to shoppers","Rank products for browsing"],"pricing":"Contact sales or per request.","status":"ga","url":"https://cloud.google.com/gemini-enterprise-cx/commerce","source":"https://cloud.google.com/products"},{"id":"gcp-food-ordering-ai-agent","provider":"gcp","name":"Food Ordering AI Agent","category":"ai-apps","tagline":"A voice and chat assistant that takes restaurant orders.","description":"Food Ordering AI Agent is a ready-made assistant that takes customers' orders by voice or chat. If you run or build tools for quick service restaurants, you could use it for drive-thru lanes or phone orders instead of building an ordering bot from scratch. It is packaged specifically for taking menu orders. It is currently in preview, which means it is still being tested and may change, and pricing is handled by contacting sales.","useCases":["Take drive-thru orders by voice","Handle phone orders for restaurants","Take menu orders in chat"],"pricing":"Contact sales.","status":"preview","url":"https://docs.cloud.google.com/food-ai","source":"https://docs.cloud.google.com/food-ordering"},{"id":"gcp-cloud-talent-solution","provider":"gcp","name":"Cloud Talent Solution","category":"specialized","also":["ai-apps"],"tagline":"Job search that understands titles, skills and places for your career site.","description":"Cloud Talent Solution is a job search tool from Google that understands what people mean when they type a job title, a skill or a location. If you run a job board or a company career page, a visitor searching for one kind of role can also see listings that use a different title for the same work, close to where they live. It can also match candidates to open roles. Pick it when you want smarter results than basic keyword search without building the matching yourself. You pay per request.","useCases":["Power job board search","Match candidates to job openings","Search jobs by title, skills, location"],"pricing":"Per request.","status":"ga","url":"https://cloud.google.com/talent-solution","source":"https://cloud.google.com/products"},{"id":"gcp-vision-api-product-search","provider":"gcp","name":"Vision API Product Search","category":"ai-apps","tagline":"Let shoppers search your catalog with a photo (retiring).","description":"Vision API Product Search lets people upload a photo and get back similar items from your product catalog. If you run an online store, a shopper could snap a picture of a lamp they like and see the closest matches you sell. It also works for suggesting similar products on a product page. Pricing is per query, so you pay as people search. This service is retiring, and Google points to AI Commerce Search as the replacement, so new projects should start there.","useCases":["Find products from a photo","Suggest visually similar products","Match shopper photos to catalog"],"pricing":"Per query.","status":"deprecated","url":"https://cloud.google.com/vision/product-search/docs","source":"https://cloud.google.com/products"},{"id":"gcp-anti-money-laundering-ai","provider":"gcp","name":"Anti Money Laundering AI","category":"specialized","also":["ai-apps"],"tagline":"Spot suspicious money movement in a bank with AI risk scores.","description":"Anti Money Laundering AI scores bank transactions so investigators can see which ones look most like money laundering. It replaces older rule-based monitoring, where fixed rules such as amount limits trigger alerts, with machine learning that ranks risk. If you build software for a bank or payments company, it helps your team review the riskiest alerts first instead of wading through piles of false alarms. It is made for financial institutions, and pricing is handled through Google's sales team.","useCases":["Flag suspicious payments for review","Rank alerts by risk level","Monitor bank transactions for laundering"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/anti-money-laundering-ai","source":"https://cloud.google.com/products"},{"id":"gcp-deep-learning-vm","provider":"gcp","name":"Deep Learning VM Image","category":"ai-platform","also":["vms"],"tagline":"Ready-made servers with AI tools and graphics chips already set up.","description":"Deep Learning VM Image is a ready-made server setup that already has popular AI tools installed, including PyTorch, TensorFlow and JAX (libraries for building and training AI models). It also includes drivers for GPUs, the graphics chips that make AI training fast. If you want to try training or fine-tuning a model this afternoon, you start one and skip hours of installing software. It is also handy for keeping your setup the same every time. The image is free, and you only pay for the server it runs on.","useCases":["Test AI models on a GPU quickly","Start with PyTorch or TensorFlow preinstalled","Repeat experiments in the same setup"],"pricing":"Free images. Pay for the VM.","status":"ga","url":"https://cloud.google.com/deep-learning-vm","source":"https://cloud.google.com/products"},{"id":"gcp-deep-learning-containers","provider":"gcp","name":"Deep Learning Containers","category":"ai-platform","also":["containers"],"tagline":"Prebuilt AI software packages that run the same on Google Cloud.","description":"Deep Learning Containers are prebuilt packages, called containers, that bundle common AI software so it runs the same way wherever you launch it. If your team trains models on GKE (Google's service for running many containers) or deploys on Cloud Run, you can start from one of these instead of fixing version clashes by hand. They work with Agent Platform too. Pick them over a single server image when you want packages that move easily between services. The images are free, and you pay for the computing power you use.","useCases":["Run AI code in consistent packaged environments","Train models on a managed cluster","Deploy AI apps on Cloud Run"],"pricing":"Free images. Pay for compute.","status":"ga","url":"https://cloud.google.com/deep-learning-containers","source":"https://cloud.google.com/products"},{"id":"gcp-compute-engine","provider":"gcp","name":"Compute Engine","category":"vms","also":["gpu"],"tagline":"Rent virtual servers from Google, including ones with AI chips.","description":"Compute Engine lets you rent virtual servers, called VMs, running on Google's hardware. If you have an app or database that needs a full server you control, you pick the size, start it and run whatever you like on it. You can choose custom sizes, and servers can include GPUs or TPUs, chips built for AI work. Billing is per second, with discounts for steady or committed use and a free tier that includes a small e2-micro server. It is the base building block for most things on Google Cloud.","useCases":["Host an app on a virtual server","Move existing servers to the cloud","Run databases on your own servers"],"pricing":"Per second. Sustained and committed use discounts. Free tier e2-micro.","status":"ga","url":"https://cloud.google.com/products/compute","source":"https://cloud.google.com/products"},{"id":"gcp-spot-vms","provider":"gcp","name":"Spot VMs","category":"vms","tagline":"Heavily discounted servers that Google can take back at short notice.","description":"Spot VMs are Compute Engine servers sold at up to 91 percent off because they use Google's spare capacity. The catch is that Google can reclaim them with short notice. If you run batch jobs, like processing a big pile of videos or data overnight, or automated test runners that check your code, a stopped job can simply restart. Pick Spot VMs over regular servers when the work can handle interruptions and you want a smaller bill. Avoid them for anything that must stay online constantly.","useCases":["Run batch jobs at lower cost","Power build servers for testing","Process data where interruptions are tolerable"],"pricing":"Spot price, a large discount to on-demand.","status":"ga","url":"https://cloud.google.com/spot-vms","source":"https://cloud.google.com/products"},{"id":"gcp-confidential-vm","provider":"gcp","name":"Confidential VM","category":"vms","also":["secrets"],"tagline":"Servers that keep your data encrypted even while it is being used.","description":"Confidential VM is a server whose memory stays encrypted while it works, using special keys built into the hardware. If your app handles sensitive data such as health or financial records, this means even Google's host machine cannot read that data while it is processed. It relies on AMD SEV and Intel TDX chip protections, and GPU options are available too. Pick it over a standard VM when rules or customers demand extra protection for data in use. It costs a small premium over the standard VM price.","useCases":["Process sensitive data in memory securely","Analyze regulated data with stronger privacy","Run joint analytics without exposing raw data"],"pricing":"Small premium over standard VM price.","status":"ga","url":"https://cloud.google.com/confidential-computing","source":"https://cloud.google.com/products"},{"id":"gcp-confidential-space","provider":"gcp","name":"Confidential Space","category":"vms","also":["secrets"],"tagline":"A locked room where partners can combine data without seeing each other's.","description":"Confidential Space is a protected environment where several companies can pool data for a shared job without handing it over to each other. If you and a partner want to analyze customer data together, or train an AI model on both of your datasets, only approved programs get to touch the data. Attestation, a check that proves the right code is running, verifies this. Pick it over a normal shared database when neither side wants to expose raw data. You pay for compute plus a small premium.","useCases":["Share data with approved programs only","Train private ML models on shared data","Verify the environment before sharing data"],"pricing":"Compute cost plus small premium.","status":"ga","url":"https://docs.cloud.google.com/confidential-computing/confidential-space/docs/confidential-space-overview","source":"https://docs.cloud.google.com/confidential-computing/confidential-space/docs/confidential-space-overview"},{"id":"gcp-shielded-vms","provider":"gcp","name":"Shielded VMs","category":"vms","also":["secrets"],"tagline":"Servers that check they have not been tampered with when they start.","description":"Shielded VMs are servers with extra protection that checks nothing has been tampered with when they start up. They use secure boot and a virtual security chip called vTPM, and they watch for changes over time. If you run a customer-facing app and want to guard against rootkits, a type of malware that hides deep inside a system, this protection is on by default for supported images. It is also useful for compliance checklists. There is no extra charge.","useCases":["Protect servers against boot-level tampering","Meet baseline security compliance requirements","Monitor server startup integrity automatically"],"pricing":"No extra charge.","status":"ga","url":"https://cloud.google.com/security/products/shielded-vm","source":"https://cloud.google.com/products"},{"id":"gcp-sole-tenant-nodes","provider":"gcp","name":"Sole-Tenant Nodes","category":"vms","tagline":"Physical servers reserved for your project alone.","description":"Sole-Tenant Nodes are physical servers used only by your project, so no other customer shares the machine. If you have software licensed to a specific machine, such as Windows or Oracle that you already own (often called bring your own license), or rules that require strict isolation, this gives you that. Node affinity labels let you control which of your virtual servers land on which machine. Pick it over regular VMs when licensing or compliance demands dedicated hardware. You pay per node, with discounts for committing.","useCases":["Run Windows or Oracle licenses you own","Keep workloads on physical servers alone","Meet compliance rules for isolation"],"pricing":"Per node, with commit discounts.","status":"ga","url":"https://cloud.google.com/compute/docs/nodes/sole-tenant-nodes","source":"https://cloud.google.com/products"},{"id":"gcp-bare-metal-solution","provider":"gcp","name":"Bare Metal Solution","category":"vms","also":["specialized"],"tagline":"Dedicated physical servers near Google Cloud, mainly for Oracle databases.","description":"Bare Metal Solution gives you whole physical servers, with no virtualization layer, placed in regional sites right next to Google Cloud. If your company runs an Oracle database or other specially licensed software that does not fit on regular virtual servers, you can keep it on this hardware while your app sits nearby on Google Cloud. Pick it when you need that kind of direct hardware access. For newer Oracle work, Google steers people to Oracle Database@Google Cloud instead. Pricing is by contacting sales.","useCases":["Run Oracle databases on dedicated hardware","Host specialized licensed software","Place servers near Google Cloud apps"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/bare-metal","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-gpus","provider":"gcp","name":"Cloud GPUs","category":"gpu","also":["vms"],"tagline":"Rent NVIDIA graphics chips to train, run and render AI work.","description":"Cloud GPUs lets you rent servers fitted with NVIDIA GPUs, the specialized chips that handle the heavy math behind AI. If you are training your own model, running a chatbot built on a large language model, or rendering video and 3D scenes, this is where you get the horsepower. Options include top-end systems such as H100, H200, B200 and GB200. You pay per GPU hour, with cheaper Spot rates or committed rates if you can plan ahead.","useCases":["Train your own AI models","Run chatbots and language models","Render graphics and 3D scenes"],"pricing":"Per GPU hour, with Spot and committed rates.","status":"ga","url":"https://cloud.google.com/gpu","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-tpu","provider":"gcp","name":"Cloud TPU","category":"gpu","also":["ai-platform"],"tagline":"Google's own AI chips for training and running large models.","description":"Cloud TPU gives you access to Tensor Processing Units, chips Google designed specifically for AI work. If you are training a large model or serving one to many users at once, you can rent them by the hour. Generations run from Trillium to Ironwood, and they can be linked into big groups called pods for larger jobs. Consider them instead of NVIDIA GPUs when you want Google's own hardware for high-volume AI, and ask about commitment discounts if your use is steady.","useCases":["Train very large AI models","Serve high-volume AI requests","Scale training across large chip pods"],"pricing":"Per chip hour, with commit discounts.","status":"ga","url":"https://cloud.google.com/tpu","source":"https://cloud.google.com/products"},{"id":"gcp-ai-hypercomputer","provider":"gcp","name":"AI Hypercomputer","category":"gpu","also":["ai-platform"],"tagline":"A ready-made blueprint for building big AI systems on Google Cloud.","description":"AI Hypercomputer is Google's recommended recipe for running AI at large scale, combining its AI chips, storage, networking and open software into one design. If you are training a very large model or serving one to lots of users, it shows how to fit the pieces together instead of figuring it out yourself. It covers both TPUs and GPUs, the two main kinds of AI chips. The blueprint itself costs nothing, and you only pay for the parts you use.","useCases":["Train frontier-scale AI models","Run AI inference efficiently at scale","Combine chips, storage and networking"],"pricing":"No charge for the architecture. Pay for components.","status":"ga","url":"https://cloud.google.com/ai-infrastructure","source":"https://cloud.google.com/products"},{"id":"gcp-cluster-director","provider":"gcp","name":"Cluster Director","category":"gpu","also":["vms"],"tagline":"Run large groups of AI chips as if they were one machine.","description":"Cluster Director, formerly called Hypercompute Cluster, sets up and watches over big groups of connected GPU or TPU machines. If you are training a large AI model across many machines, it helps you start the cluster and keep an eye on its health. It works with Slurm, a common job scheduler for shared computing clusters, or Kubernetes, a popular system for running apps across many machines. There is no separate fee, since you just pay for the underlying computing.","useCases":["Manage large GPU clusters as one","Monitor health of AI training clusters","Set up clusters with Slurm or Kubernetes"],"pricing":"Pay for underlying compute.","status":"ga","url":"https://docs.cloud.google.com/cluster-director/docs/overview","source":"https://docs.cloud.google.com/cluster-director/docs/overview"},{"id":"gcp-cluster-toolkit","provider":"gcp","name":"Cluster Toolkit","category":"devtools","also":["gpu"],"tagline":"Free open source kit for setting up AI and supercomputing clusters.","description":"Cluster Toolkit is a free, open source set of ready-made templates for building computing clusters on Google Cloud. If your team runs heavy scientific or AI jobs, often called HPC for high-performance computing, you can launch a cluster from these blueprints instead of wiring it by hand. They are built on Terraform, a tool that sets up cloud resources from written instructions, and cover Slurm, a popular job scheduler, plus other setups. The toolkit costs nothing, and you pay only for the resources it creates.","useCases":["Deploy high-performance computing clusters","Set up Slurm clusters quickly","Reuse open source cluster blueprints"],"pricing":"Free. Pay for resources.","status":"ga","url":"https://docs.cloud.google.com/cluster-toolkit/docs/overview","source":"https://docs.cloud.google.com/cluster-toolkit/docs/overview"},{"id":"gcp-batch","provider":"gcp","name":"Batch","category":"workflows","also":["vms"],"tagline":"Submit a job and Google runs it on temporary servers, then cleans up.","description":"Batch is a job runner: you hand it a big task and it starts the servers, runs the work, then shuts everything down. If you need to process a pile of data, render a set of video frames, or run a scientific simulation, you do not have to babysit any machines. It is fully managed, so Google handles the scheduling. You pay only for the servers and disks used, which suits work that happens now and then better than keeping servers running all the time.","useCases":["Run scheduled compute jobs automatically","Render large batches of images","Process large batches of data"],"pricing":"Pay for the VMs and disks used.","status":"ga","url":"https://cloud.google.com/batch","source":"https://cloud.google.com/products"},{"id":"gcp-vm-manager","provider":"gcp","name":"VM Manager","category":"management","also":["vms"],"tagline":"Keep a fleet of Google Cloud servers patched, listed and in line.","description":"VM Manager helps you look after many Google Cloud servers, also called VMs, at once. If you run a growing number of servers for your product, you can use it to roll out security updates, known as patches, on a schedule such as monthly, and to see what software is installed on each one. It also checks that settings stay as you intended, on both Linux and Windows. Choose it over patching by hand once the servers get hard to track. Pricing is per server per month beyond a free tier.","useCases":["Patch servers on a monthly schedule","Track software installed across servers","Keep server configurations consistent"],"pricing":"Per VM per month beyond free tier.","status":"ga","url":"https://docs.cloud.google.com/compute/docs/vm-manager","source":"https://docs.cloud.google.com/compute/docs/vm-manager"},{"id":"gcp-workload-manager","provider":"gcp","name":"Workload Manager","category":"management","tagline":"Checks your SAP and other workloads against best-practice rules.","description":"Workload Manager scans your setup against a list of recommended rules and warns you when something drifts away from them. If your company runs SAP, business software often used for finance and operations, you can use it to check that the system is configured the way experts advise. It also handles general configuration checks, so you can catch unwanted changes early. Pick it when you want automatic, repeatable checks instead of reviewing settings by hand. Pricing is based on usage.","useCases":["Check SAP systems against best practices","Catch configuration drift in workloads","Validate workloads against set rules"],"pricing":"Usage-based.","status":"ga","url":"https://docs.cloud.google.com/workload-manager/docs/overview","source":"https://docs.cloud.google.com/workload-manager/docs/overview"},{"id":"gcp-capacity-planner","provider":"gcp","name":"Capacity Planner","category":"management","tagline":"See your past computing use and forecast what you will need.","description":"Capacity Planner shows how much computing you have used over time and forecasts how much you will need. If you are about to reserve servers ahead of a launch or busy season, it helps you decide how much to reserve. It works across projects, the separate workspaces in your Google Cloud account. It is free, so it is an easy first stop before you commit to reservations.","useCases":["Forecast future server usage","Plan reserved capacity ahead of time","Review usage across all projects"],"pricing":"No charge.","status":"ga","url":"https://docs.cloud.google.com/capacity-planner/docs/overview","source":"https://docs.cloud.google.com/capacity-planner/docs/overview"},{"id":"gcp-container-optimized-os","provider":"gcp","name":"Container-Optimized OS","category":"vms","also":["containers"],"tagline":"A slim, Google-maintained operating system for running containers.","description":"Container-Optimized OS is a small Linux operating system, the base software a server runs on, built for running containers. Containers are packaged apps that run the same way anywhere, typically with Docker. If you are running your app in a container on a single server, or on the servers in a GKE cluster (GKE is Google's managed Kubernetes service), this is a ready-made starting point. Google maintains it and it updates automatically. The image is free.","useCases":["Run container hosts with automatic updates","Launch single-container virtual machines","Power nodes for Kubernetes clusters"],"pricing":"Free image.","status":"ga","url":"https://docs.cloud.google.com/container-optimized-os/docs","source":"https://docs.cloud.google.com/container-optimized-os/docs"},{"id":"gcp-vmware-engine","provider":"gcp","name":"Google Cloud VMware Engine","category":"vms","also":["migration"],"tagline":"Run your existing VMware setup inside Google Cloud.","description":"Google Cloud VMware Engine lets you run the VMware environment your company already uses inside Google Cloud. If you are closing a data center or want a backup site for disaster recovery, you can move your setup over as it is, known as lift-and-shift, without rebuilding your apps. It runs on dedicated physical servers with vSphere, vSAN and NSX, VMware's tools for virtual machines, storage and networking. Pick it over rewriting everything for the cloud when speed of moving matters. Pricing is per node, with discounts for commitments.","useCases":["Move VMware setups into Google Cloud","Run disaster recovery for VMware systems","Exit your own data center"],"pricing":"Per node, with commit discounts.","status":"ga","url":"https://cloud.google.com/vmware-engine","source":"https://cloud.google.com/products"},{"id":"gcp-oracle-database-at-google-cloud","provider":"gcp","name":"Oracle Database@Google Cloud","category":"sql","tagline":"Oracle's Exadata and Autonomous Database, hosted in Google data centers.","description":"Oracle Database@Google Cloud places Oracle's database systems inside Google Cloud data centers. That includes Exadata, Oracle's high-end database hardware, and Autonomous Database, Oracle's self-managing database. Oracle runs the hardware, and it connects to your Google Cloud apps with low delay. If your company depends on Oracle and is moving the rest of its stack to Google, you can keep Oracle without leaving the database far from your apps. Billing goes through Oracle, and it is listed in Marketplace.","useCases":["Migrate Oracle databases to Google Cloud","Run Exadata near your Google apps","Connect Oracle databases to apps with low latency"],"pricing":"Billed via Oracle. See Marketplace.","status":"ga","url":"https://cloud.google.com/solutions/oracle","source":"https://cloud.google.com/products"},{"id":"gcp-oracle-on-google-cloud-compute","provider":"gcp","name":"Oracle on Google Cloud Compute","category":"vms","tagline":"Run your own Oracle databases on Google's servers, with setup guidance.","description":"This lets you run Oracle, the database software many large companies depend on, on Google Cloud servers that you manage yourself. If your business already runs Oracle in its own data center and wants to move it without rewriting anything, this is the route, often called lift-and-shift. It also covers Oracle RAC, a setup that spreads one database across several machines, and it can use bare metal, meaning whole physical machines. Pick it when you need to keep Oracle exactly as it is. You pay for the infrastructure and for the Oracle licenses.","useCases":["Move Oracle workloads onto virtual servers","Run self-managed Oracle RAC clusters","Lift existing Oracle systems to the cloud"],"pricing":"Pay for infrastructure and licenses.","status":"ga","url":"https://docs.cloud.google.com/oracle/compute/overview","source":"https://docs.cloud.google.com/compute/docs/oracle"},{"id":"gcp-sap-on-google-cloud","provider":"gcp","name":"SAP on Google Cloud","category":"specialized","also":["vms"],"tagline":"Google-certified servers and tools for running SAP business software.","description":"This is Google Cloud's setup for running SAP, the business software many companies use for finance, orders and inventory. It includes certified virtual servers, bare metal (whole physical machines) and tooling for SAP HANA, SAP's database, and S/4HANA, its main ERP system. If your company is moving its SAP systems out of its own data center, this gives you hardware that is certified for SAP workloads. Choose it over general-purpose servers when SAP needs to run on approved infrastructure. You pay for the infrastructure you use.","useCases":["Host SAP HANA on certified infrastructure","Migrate ERP systems to Google Cloud","Run S/4HANA on certified servers"],"pricing":"Pay for infrastructure.","status":"ga","url":"https://cloud.google.com/solutions/sap","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-run","provider":"gcp","name":"Cloud Run","category":"serverless","also":["containers"],"tagline":"Run your app from a container and pay only while it handles requests.","description":"Cloud Run runs your app from a container, which is a packaged copy of your app with everything it needs to run. You can also deploy from a source code repository or a single function. If you're building an API for your product or a job that runs in the background, you hand over the code and Google handles the servers. When nobody is using it, it scales down to zero, so you aren't paying for idle time. It also supports GPUs and worker pools, and the free tier covers 2 million requests per month.","useCases":["Deploy HTTP APIs from any container","Run background jobs that scale to zero","Host backends for AI agents"],"pricing":"Usage-based. Free tier: 2M requests per month.","status":"ga","url":"https://cloud.google.com/run","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-run-functions","provider":"gcp","name":"Cloud Run functions","category":"serverless","tagline":"Small pieces of code that run when something happens, like a webhook.","description":"Cloud Run functions lets you write a small function and have Google run it whenever something triggers it. If another service sends your app a webhook, which is an automatic message about an event, a function can catch it and respond. It can also react to a new file landing in storage or a message from Pub/Sub, Google's messaging service. It is the new name for Cloud Functions 2nd gen and runs on Cloud Run, so it shares the same pricing model and a free monthly tier.","useCases":["Run code when webhooks arrive","React to files uploaded to storage","Handle messages from a message queue"],"pricing":"Same model as Cloud Run. Free monthly tier.","status":"ga","url":"https://cloud.google.com/functions","source":"https://cloud.google.com/products"},{"id":"gcp-app-engine","provider":"gcp","name":"App Engine","category":"hosting","also":["serverless"],"tagline":"Google's original app hosting, which grows with your traffic automatically.","description":"App Engine is Google's long-running service for hosting web apps without managing servers. You upload your code in Python, Java, Node.js, Go, PHP or Ruby, and it adds capacity as visitors arrive. If you're keeping an older web app alive or running a simple site that needs to scale, it still does the job well. It is mature, but newer apps usually pick Cloud Run instead. You pay for instance hours, and the standard environment includes a free daily quota.","useCases":["Host older web apps","Run simple sites that scale automatically","Deploy Python, Java, Node.js or Go apps"],"pricing":"Instance hours. Free daily quota in standard.","status":"ga","url":"https://cloud.google.com/appengine","source":"https://cloud.google.com/products"},{"id":"gcp-google-kubernetes-engine","provider":"gcp","name":"Google Kubernetes Engine (GKE)","category":"containers","tagline":"Google-run Kubernetes for running many containers at scale.","description":"Kubernetes is a popular system for running and coordinating many containers, which are packaged copies of your app. Google Kubernetes Engine (GKE) is Google's managed version, so you don't build that system yourself. If your product is split into many small services, or you need to serve AI models on GPUs, GKE can run them together. In Autopilot mode Google manages the underlying machines for you, while Standard mode gives you full control of them. You pay an hourly cluster fee plus the resources you use, and a free tier credit applies.","useCases":["Run many small connected services","Serve AI models on GPUs","Run apps that keep their own data"],"pricing":"Cluster fee per hour plus node or pod resources. Free tier credit.","status":"ga","url":"https://cloud.google.com/kubernetes-engine","source":"https://cloud.google.com/products"},{"id":"gcp-gke-enterprise","provider":"gcp","name":"GKE Enterprise","category":"containers","also":["management"],"tagline":"Manage Kubernetes across clouds and your own servers, formerly Anthos.","description":"GKE Enterprise helps teams that run Kubernetes, a system for coordinating containers, in more than one place. It adds shared policies, a service mesh (a layer that controls how services talk), config syncing and multi-cluster tools to GKE, GKE on AWS and Azure, and clusters you attach from elsewhere. If your company runs apps in several clouds and its own data center, you can manage them from one spot. It was previously called Anthos. Pricing is per vCPU per hour, where a vCPU is a unit of computing power.","useCases":["Manage many Kubernetes clusters together","Run clusters across clouds and own data centers"],"pricing":"Per vCPU per hour.","status":"ga","url":"https://docs.cloud.google.com/kubernetes-engine/enterprise/docs/deployment-options","source":"https://cloud.google.com/products"},{"id":"gcp-config-sync","provider":"gcp","name":"Config Sync","category":"devtools","also":["containers"],"tagline":"Keep Kubernetes clusters matching the settings stored in a Git repo.","description":"Config Sync keeps your Kubernetes clusters set up exactly the way a Git repository says they should be. Git is the tool most teams use to track changes to files. If someone edits a cluster by hand, Config Sync notices and brings it back in line, and a change approved in Git rolls out to every cluster. This approach is called GitOps. You can also use an OCI source, a standard way of packaging files, instead of Git. It is included with GKE Enterprise.","useCases":["Keep clusters matching a Git repo","Push one config to every cluster"],"pricing":"Included with GKE Enterprise.","status":"ga","url":"https://docs.cloud.google.com/kubernetes-engine/enterprise/config-sync/docs/overview","source":"https://docs.cloud.google.com/kubernetes-engine/enterprise/config-sync/docs/overview"},{"id":"gcp-policy-controller","provider":"gcp","name":"Policy Controller","category":"management","also":["containers"],"tagline":"Set safety rules for your Kubernetes clusters and check they're followed.","description":"Policy Controller lets you write rules for what is allowed inside your Kubernetes clusters, and it enforces them automatically. It is built on OPA Gatekeeper, a widely used open-source rule engine. If you want to block risky settings, such as apps that run with too many permissions, it stops them before they start. It also audits what is already running, which helps when you need to show compliance. It works across a whole fleet of clusters and is included with GKE Enterprise.","useCases":["Block risky container settings automatically","Check clusters against compliance rules"],"pricing":"Included with GKE Enterprise.","status":"ga","url":"https://docs.cloud.google.com/kubernetes-engine/enterprise/policy-controller/docs/overview","source":"https://docs.cloud.google.com/kubernetes-engine/enterprise/policy-controller/docs/overview"},{"id":"gcp-config-connector","provider":"gcp","name":"Config Connector","category":"devtools","also":["containers"],"tagline":"Create Google Cloud resources by writing Kubernetes config files.","description":"Config Connector is an add-on for Kubernetes, a system for coordinating containers, that lets you create Google Cloud resources from config files. If your team already describes its apps in Kubernetes YAML, a simple text format for settings, you can describe things like databases and storage there too. That keeps your apps and the cloud services they rely on in one place and in sync. It also fits GitOps, where changes to files in Git drive changes to your setup. The add-on itself is free.","useCases":["Create cloud resources from Kubernetes files","Describe cloud resources in config files"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/config-connector/docs/overview","source":"https://cloud.google.com/products"},{"id":"gcp-config-controller","provider":"gcp","name":"Config Controller","category":"devtools","also":["containers"],"tagline":"A Google-hosted control center that manages your cloud setup from config files.","description":"Config Controller is a Kubernetes cluster that Google hosts for you, already loaded with Config Sync and Config Connector. Together they let you describe your cloud resources in files and keep everything matching those files automatically. If a platform team wants one central place to manage settings for many projects, this saves you from building that control point yourself. Pick it over installing the two tools by hand when you'd rather have Google run the setup. You pay the cost of the cluster.","useCases":["Run a managed control plane for config","Manage cloud resources from one hosted setup","Run GitOps for cloud resources"],"pricing":"Cluster cost.","status":"ga","url":"https://docs.cloud.google.com/kubernetes-engine/enterprise/config-controller/docs/overview","source":"https://docs.cloud.google.com/kubernetes-engine/enterprise/config-controller/docs/overview"},{"id":"gcp-cloud-service-mesh","provider":"gcp","name":"Cloud Service Mesh","category":"networking","also":["containers"],"tagline":"Control and secure how your app's services talk to each other.","description":"Cloud Service Mesh is a managed service, based on Istio, that sits between the parts of your app and manages how they communicate. If your product is split into many small services, it can encrypt their traffic with mTLS, which makes both sides prove who they are. It can also send a small share of visitors to a new version before a full rollout, known as a canary release. It gives you visibility into traffic too. It works with GKE, Cloud Run and VMs, and is priced per client workload hour or through GKE Enterprise.","useCases":["Encrypt traffic between your services","Split traffic for gradual rollouts","Track traffic across services"],"pricing":"Per client workload hour or via GKE Enterprise.","status":"ga","url":"https://cloud.google.com/products/service-mesh","source":"https://cloud.google.com/products"},{"id":"gcp-knative-serving","provider":"gcp","name":"Knative serving","category":"serverless","also":["containers"],"tagline":"Run apps that scale with demand inside your own Kubernetes cluster.","description":"Knative serving is Google's version of Knative, an open source tool for running apps in a serverless way, where the app grows and shrinks with traffic instead of you managing each server. It runs on GKE, Google's managed Kubernetes, which is a system for running packaged apps across many machines. If your team already has a GKE cluster and wants apps that respond to demand, this is how you get that. Pick it over Cloud Run when you want portable serverless apps that stay inside your own cluster. You pay the cluster cost.","useCases":["Run serverless apps inside your cluster","Keep serverless apps portable on GKE"],"pricing":"Cluster cost.","status":"ga","url":"https://docs.cloud.google.com/run/docs/overview/what-is-cloud-run","source":"https://docs.cloud.google.com/run/docs/overview/what-is-cloud-run"},{"id":"gcp-artifact-registry","provider":"gcp","name":"Artifact Registry","category":"devtools","also":["containers"],"tagline":"A private place to store your app's container images and code packages.","description":"Artifact Registry is private storage for the built pieces of your software, such as Docker container images (packaged apps ready to run) and code packages for npm, Maven, Python and Go. If your team turns an app into a container and needs somewhere safe to keep it before deploying, this is where it goes. It also offers scanning of what you store. Choose it over the older Container Registry, which it replaced. You pay per GB stored plus egress, meaning data leaving Google's network.","useCases":["Store container images for deployments","Host private npm, Maven and Python packages","Scan stored images for security issues"],"pricing":"Per GB storage plus egress.","status":"ga","url":"https://cloud.google.com/artifact-registry","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-build","provider":"gcp","name":"Cloud Build","category":"devtools","tagline":"Automatically build and test your code every time you push changes.","description":"Cloud Build builds, tests and packages your code automatically, with no build server for you to look after. If you push changes to GitHub, GitLab or Bitbucket and want your app tested and turned into a container each time, you set up a trigger and Cloud Build does the rest. You list the steps in a YAML file, a simple settings file, and each step runs in its own container. It is a natural pick if you already host your app on Google Cloud. You pay per build minute, and the first 2,500 minutes each month are free.","useCases":["Build containers from new GitHub commits","Run automated build and test pipelines"],"pricing":"Per build minute. Free 2,500 minutes per month.","status":"ga","url":"https://cloud.google.com/build","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-deploy","provider":"gcp","name":"Cloud Deploy","category":"devtools","also":["containers"],"tagline":"Move releases through test and live stages with approvals and rollbacks.","description":"Cloud Deploy automates how a new version of your app travels from testing to staging to live. If you run on GKE (Google's managed Kubernetes), Cloud Run or Agent Runtime, you set up a pipeline of stages and the tool carries each release forward. You can require a person to approve before a release reaches real users, and roll back if something breaks. Pick it when pushing releases by hand has become risky or messy. You pay per active delivery pipeline.","useCases":["Roll out releases in stages","Require approval before each release","Roll back a bad release"],"pricing":"Per active delivery pipeline.","status":"ga","url":"https://cloud.google.com/deploy","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-code","provider":"gcp","name":"Cloud Code","category":"devtools","tagline":"Add Google Cloud tools to your code editor for building and deploying apps.","description":"Cloud Code is a free set of add-ons for VS Code and JetBrains, two popular code editors. If you are building an app meant for Cloud Run or Kubernetes, it lets you run and debug it on your own computer using emulators, which are stand-ins for cloud services, and then deploy without leaving the editor. It saves you from juggling the editor and a pile of separate command-line steps. Choose it if you already work in one of those editors. It costs nothing.","useCases":["Debug Kubernetes apps inside VS Code","Deploy Cloud Run apps from your editor","Test apps locally with emulators"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/code","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-workstations","provider":"gcp","name":"Cloud Workstations","category":"devtools","also":["end-user"],"tagline":"A ready-to-code computer in the cloud that you open from your browser.","description":"Cloud Workstations gives you a development computer that lives in the cloud instead of on your laptop. If you are bringing on a new teammate or contractor, they can open a ready setup in a browser or code editor without spending days installing tools. Admins can add security controls and custom images, which are preset bundles of software. Pick it over personal laptops when you want consistent, locked-down setups for your code. You pay for the virtual computer and disk time while you use them.","useCases":["Give developers secure cloud dev machines","Set up new developers faster","Use custom machine images for teams"],"pricing":"VM and disk time.","status":"ga","url":"https://cloud.google.com/workstations","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-shell","provider":"gcp","name":"Cloud Shell","category":"devtools","tagline":"A free command-line window in your browser, with tools already set up.","description":"Cloud Shell is a terminal, the text window where you type commands, that opens right in your browser. It runs Debian, a common version of Linux, and comes with gcloud (Google Cloud's command tool), an editor and Docker already installed. If you are following a tutorial or need to run a quick admin task, you skip installing anything on your own computer. Files in your home folder stick around between sessions, with 5 GB of space. It is free.","useCases":["Run admin commands from the browser","Follow cloud tutorials in a terminal","Keep files in a free home folder"],"pricing":"Free with 5 GB home disk.","status":"ga","url":"https://cloud.google.com/shell","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-sdk","provider":"gcp","name":"gcloud CLI and Cloud SDK","category":"devtools","tagline":"Control Google Cloud by typing commands or writing scripts.","description":"The gcloud CLI and Cloud SDK are a free toolkit for managing Google Cloud with typed commands instead of clicking through the website. It includes gcloud, bq (for BigQuery data) and gsutil (for Cloud Storage files), plus client libraries, which are code add-ons that let your own app talk to Google Cloud. If you find yourself repeating the same setup steps, a script is far faster than clicking. Choose it when you outgrow the web console and want to automate. It costs nothing.","useCases":["Script cloud tasks from the command line","Automate Google Cloud tasks","Call cloud services from code libraries"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/sdk","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-source-repositories","provider":"gcp","name":"Cloud Source Repositories","category":"devtools","tagline":"Private Git code hosting from Google, now closed to new customers.","description":"Cloud Source Repositories hosted private Git repositories, which are folders that track every change to your code. If you already have code stored there, it keeps working for you, and people also used it to mirror repos that live elsewhere. For new projects, Google suggests GitHub, GitLab or Secure Source Manager instead. It has been closed to new customers since 2024, so only existing customers can use it today.","useCases":["Keep existing private Git repos running","Mirror repos you already host","Move code to GitHub or GitLab"],"pricing":"Existing customers only.","status":"closed","url":"https://cloud.google.com/source-repositories/docs","source":"https://cloud.google.com/products"},{"id":"gcp-secure-source-manager","provider":"gcp","name":"Secure Source Manager","category":"devtools","tagline":"Your own private home for code, with issues and code review.","description":"Secure Source Manager is a private place to store and review your code, running as your own instance that is not shared with other customers. It includes issues for tracking tasks and pull requests, which are proposed changes your teammates review before they are merged. If you work in a regulated field and cannot keep code on a shared public service, this fits well. Pick it over a service like GitHub when you need that separation. You pay per instance hour.","useCases":["Host private code in your own instance","Review code changes before merging","Keep source code in regulated settings"],"pricing":"Per instance hour.","status":"ga","url":"https://docs.cloud.google.com/secure-source-manager/docs/overview","source":"https://docs.cloud.google.com/secure-source-manager/docs/overview"},{"id":"gcp-developer-connect","provider":"gcp","name":"Developer Connect","category":"devtools","tagline":"Link your GitHub or GitLab code to Google Cloud tools.","description":"Developer Connect links your code on sites like GitHub and GitLab to Google Cloud. If you want Cloud Build to run automatically each time you push code, this is the connection that makes it possible. It also lets Gemini code tools see your repositories. Pick it when your code lives outside Google and you want to keep it there instead of moving it. There is no charge for using it.","useCases":["Link GitHub repos to Cloud Build","Connect GitLab and other code hosts","Give Gemini code tools access to repos"],"pricing":"No charge.","status":"ga","url":"https://docs.cloud.google.com/developer-connect/docs/overview","source":"https://docs.cloud.google.com/developer-connect/docs/overview"},{"id":"gcp-infrastructure-manager","provider":"gcp","name":"Infrastructure Manager","category":"devtools","tagline":"Set up your cloud with Terraform files, run for you by Google.","description":"Infrastructure Manager runs Terraform for you. Terraform is a popular way to describe your cloud setup, such as servers, databases and networks, in files so you can recreate it the same way each time. If you want a standard stack that your whole team deploys identically, you hand over your files and it runs them, keeping its record of what exists in Cloud Storage. It replaces Deployment Manager, Google's older tool for this job. You pay for Terraform execution minutes beyond a free tier.","useCases":["Run Terraform deployments for you","Keep Terraform state in Cloud Storage","Standardize cloud setups across teams"],"pricing":"Terraform execution minutes beyond free tier.","status":"ga","url":"https://cloud.google.com/infrastructure-manager/docs","source":"https://cloud.google.com/products"},{"id":"gcp-deployment-manager","provider":"gcp","name":"Deployment Manager","category":"devtools","tagline":"Older Google tool that sets up cloud resources from text files.","description":"Deployment Manager is an older Google Cloud tool that builds your servers, databases and networks from a text file describing what you want, written in YAML, a simple settings format. If you already have templates running your setup, this is what they use, and Google's replacement is Infra Manager. For anything new, start with Infra Manager instead, since new users are blocked from Deployment Manager. Support ended in April 2026, new users have been blocked since June 30, 2026, and the service turns down on June 30, 2027.","useCases":["Keep existing templates running until 2027","Move templates to Infrastructure Manager"],"pricing":"Free; being retired.","status":"deprecated","url":"https://cloud.google.com/deployment-manager/docs/deprecations","source":"https://cloud.google.com/products"},{"id":"gcp-terraform-on-google-cloud","provider":"gcp","name":"Terraform on Google Cloud","category":"devtools","tagline":"Official Terraform tools and ready-made blueprints for Google Cloud.","description":"Terraform on Google Cloud is Google's official support for Terraform, a popular tool that creates your cloud setup from files instead of clicking through menus. If you want the test version and live version of your app to match exactly, you write the setup once and reuse it. Google also publishes ready-made blueprints, including starting layouts for a company's cloud account (often called landing zones). Pick this over a hand-made setup when you want something repeatable that Google itself supports. It is free to use.","useCases":["Write cloud setup as Terraform code","Start from ready-made landing zone blueprints","Use official Google Cloud Terraform providers"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/docs/terraform","source":"https://docs.cloud.google.com/docs/terraform"},{"id":"gcp-application-design-center","provider":"gcp","name":"Application Design Center","category":"devtools","also":["management"],"tagline":"Draw your app's cloud setup on a canvas, then export it as Terraform.","description":"Application Design Center is a drawing board where you piece together the parts of an app, such as a website, a database and file storage, with Gemini, Google's AI assistant, helping with the design. When you are done, you export it as Terraform, a popular tool that builds cloud setups from files. If you are planning a new product and are not sure how the pieces should connect, this lets you see it before you build it. It is in preview, which means it is still early and may change.","useCases":["Sketch app architecture on a visual canvas","Export designs as Terraform code","Build reference architectures for teams"],"pricing":"Preview pricing; see page.","status":"preview","url":"https://docs.cloud.google.com/application-design-center/docs/overview","source":"https://docs.cloud.google.com/application-design-center/docs/overview"},{"id":"gcp-app-hub","provider":"gcp","name":"App Hub","category":"management","tagline":"See your cloud pieces grouped by the app they belong to.","description":"App Hub lets you group the services and programs you run in Google Cloud into named apps. If your product has a website, a database and a few background jobs scattered across different places, you can bundle them into one app, such as \"checkout\", and see who owns it and how it is doing. Normal cloud screens list things by type, while App Hub organizes them by the product they serve, which is easier to follow as your setup grows. There is no charge for it.","useCases":["Group services into one app view","Track ownership of cloud resources","See an inventory of your apps"],"pricing":"No charge.","status":"ga","url":"https://docs.cloud.google.com/app-hub/docs/overview","source":"https://docs.cloud.google.com/app-hub/docs/overview"},{"id":"gcp-app-lifecycle-manager","provider":"gcp","name":"App Lifecycle Manager","category":"devtools","also":["management"],"tagline":"Roll out updates to software you run for many customers.","description":"App Lifecycle Manager helps companies that deliver software as a managed service, where each customer gets their own running copy. If you run your own SaaS product (software people subscribe to) and need to ship a new version to many customer setups, you can plan and track those rollouts in one place. That beats updating each customer by hand, and it keeps track of which version everyone is on. It is in preview, which means it is still early and may change.","useCases":["Roll out new versions to each customer","Manage deployments of a hosted app","Control which app version each customer runs"],"pricing":"See product page.","status":"preview","url":"https://cloud.google.com/products/app-lifecycle-manager","source":"https://docs.cloud.google.com/app-lifecycle-manager/docs"},{"id":"gcp-buildpacks","provider":"gcp","name":"Google Cloud Buildpacks","category":"devtools","also":["containers"],"tagline":"Turn your code into a runnable container without writing a Dockerfile.","description":"Google Cloud Buildpacks look at your code, work out what language it uses, and package it into a container, which is a ready-to-run bundle that behaves the same everywhere. If an AI tool gave you a Python or Node app and you want it online, you can deploy straight from your source code to Cloud Run, Google's service for running apps. The alternative is writing a Dockerfile, a technical recipe file, by hand, and buildpacks let you skip that. They are open source, used by Cloud Run and Cloud Build, and free.","useCases":["Turn source code into a container","Ship code to Cloud Run without Docker","Build container images with no Dockerfile"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/docs/buildpacks/overview","source":"https://docs.cloud.google.com/docs/buildpacks/overview"},{"id":"gcp-service-catalog","provider":"gcp","name":"Service Catalog","category":"management","tagline":"A shortlist of approved cloud setups your team can pick from.","description":"Service Catalog is a private menu of cloud setups that your company has already approved. An admin publishes ready-made solutions built with Terraform, a popular tool that creates cloud resources from files, and teammates pick one when they need it. If you want everyone to launch a database or app the same safe way without asking an admin each time, this does that. It suits teams with shared rules more than a solo builder. It is free.","useCases":["Share approved templates with internal teams","Give staff a company app store","Publish vetted setups for team use"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/service-catalog/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-marketplace","provider":"gcp","name":"Google Cloud Marketplace","category":"management","tagline":"Buy and set up outside software, billed through your Google Cloud account.","description":"Google Cloud Marketplace is a catalog of software from other companies that you can buy and set up without leaving Google Cloud. It includes subscription apps (SaaS), virtual servers and containers, which are ready-to-run app bundles. If you need a monitoring tool or database from another vendor, you can deploy it in a few clicks. The big difference is billing: charges land on your Google Cloud bill and can count toward a spending commitment you have already agreed with Google. Prices are set by each vendor.","useCases":["Buy third-party software on your Google bill","Deploy vendor tools with a few clicks","Use existing Google commitment spend"],"pricing":"Vendor pricing.","status":"ga","url":"https://cloud.google.com/marketplace","source":"https://cloud.google.com/products"},{"id":"gcp-bigquery","provider":"gcp","name":"BigQuery","category":"warehouse","also":["ai-platform"],"tagline":"Ask questions of huge piles of data using SQL, with no servers to run.","description":"BigQuery is a place to store large amounts of data and ask questions about it using SQL, a common language for working with data. If you want to know which marketing channels bring paying users out of millions of event records, you can run that query in seconds. There are no servers to manage, and it can also run machine learning and search by meaning right inside your queries. You pay per amount of data scanned, or reserve capacity instead. The first 1 TB of queries each month is free.","useCases":["Run analytics on very large datasets","Train machine learning models with SQL","Search and analyze application logs"],"pricing":"On-demand per TB scanned or capacity slots. Free 1 TB queries per month.","status":"ga","url":"https://cloud.google.com/bigquery","source":"https://cloud.google.com/products"},{"id":"gcp-lakehouse","provider":"gcp","name":"Lakehouse for Apache Iceberg","category":"warehouse","also":["block-storage"],"tagline":"Keep data in open table files that BigQuery, Spark and others can all read.","description":"Lakehouse for Apache Iceberg, formerly BigLake, manages Iceberg tables, an open way of storing big tables of data as files. If your data team uses several tools, such as BigQuery, Spark, Trino and Flink, they can all work on the same tables without making copies. It also includes a shared catalog, which is a directory that tells each tool where the data lives. Choose it over a single closed warehouse when you want to avoid being locked in. It can even query data stored in Amazon S3. You pay for storage plus the tools you use.","useCases":["Query Iceberg tables from several engines","Analyze data stored in AWS S3","Keep data in an open format"],"pricing":"Storage plus engine usage.","status":"ga","url":"https://cloud.google.com/products/lakehouse","source":"https://cloud.google.com/products"},{"id":"gcp-knowledge-catalog","provider":"gcp","name":"Knowledge Catalog","category":"data-pipelines","also":["ai-search"],"tagline":"Find and trust your company data, and give AI agents the context.","description":"Knowledge Catalog, formerly Dataplex, is a searchable index of the data your company has. It gathers details from BigQuery, AlloyDB, Spanner and other databases, then adds quality checks, lineage (a record of where data came from) and meaning that AI agents can use. If your team cannot tell which table holds the real revenue numbers, this helps you find and trust the right one. It goes beyond a plain list by tying governance and context together. Pricing is based on usage, with a free tier for the basic catalog.","useCases":["Find and document company data assets","Track data quality and lineage","Give AI agents business context"],"pricing":"Usage-based; free tier for basic catalog.","status":"ga","url":"https://cloud.google.com/products/knowledge-catalog","source":"https://cloud.google.com/products"},{"id":"gcp-data-catalog","provider":"gcp","name":"Data Catalog","category":"data-pipelines","tagline":"Older data-finding tool, replaced by Knowledge Catalog.","description":"Data Catalog is an older Google Cloud tool for searching and labeling information about your data, such as which tables exist and what they contain. If you already use it to tag your data, plan to move those tags to Knowledge Catalog, which Google now recommends for all new work. New projects should start there instead. Data Catalog is scheduled to be discontinued in 2026 and is available to existing customers only.","useCases":["Keep existing tag-based search running","Move existing tags to the new catalog","Search metadata in the old catalog"],"pricing":"Existing customers only.","status":"deprecated","url":"https://docs.cloud.google.com/data-catalog/docs/release-notes","source":"https://docs.cloud.google.com/data-catalog/docs/release-notes"},{"id":"gcp-looker","provider":"gcp","name":"Looker","category":"warehouse","also":["analytics"],"tagline":"Dashboards and reports built on one shared set of business definitions.","description":"Looker is a business intelligence tool, meaning it turns your company's data into dashboards and reports. Its modeling language, LookML, lets you define terms like \"revenue\" or \"active customer\" once, so every chart gives the same answer. If you run a software product and want customers to see charts inside your app, Looker supports that through embedded analytics. It also includes Gemini agents, Google's AI helpers. Pick it over a free dashboard tool when many teams need to trust the same numbers. Pricing is by contacting sales, based on the platform plus users.","useCases":["Define one shared set of metrics","Embed dashboards inside your product","Give teams governed reports and dashboards"],"pricing":"Contact sales; platform plus users.","status":"ga","url":"https://cloud.google.com/looker","source":"https://cloud.google.com/products"},{"id":"gcp-data-studio","provider":"gcp","name":"Data Studio","category":"warehouse","also":["analytics"],"tagline":"Free dashboards and reports you can build and share with your team.","description":"Data Studio is Google's free tool for turning your data into charts, dashboards and reports. It was called Looker Studio, and was renamed back to Data Studio in April 2026. If you're a marketer who wants a weekly report on ad results, you can build it once and share a link with your team. Pick it over Looker when you want quick, one-off reports without a sales call or a big budget. A Pro tier, priced per user per month, adds options for teams that need more.","useCases":["Build free dashboards for campaigns","Share marketing performance reports with clients","Create quick ad hoc reports"],"pricing":"Free. Pro per user per month.","status":"ga","url":"https://cloud.google.com/data-studio","source":"https://cloud.google.com/products"},{"id":"gcp-dataflow","provider":"gcp","name":"Dataflow","category":"data-pipelines","tagline":"Run jobs that clean and move data, in batches or as it arrives.","description":"Dataflow is a managed service that runs jobs to clean up, reshape and move data, either in big batches or continuously as new data arrives. It runs Apache Beam, an open source framework for writing those jobs. If your app produces a stream of events like clicks or orders, you can use Dataflow to enrich each one and load it into a database as it happens. It scales up and down on its own and offers ready-made templates, so you don't start from a blank page. Pick it when you need real time results instead of nightly jobs.","useCases":["Process data streams as they arrive","Enrich events in real time","Run scheduled batch data jobs"],"pricing":"Per vCPU, memory and data hour.","status":"ga","url":"https://cloud.google.com/products/dataflow","source":"https://cloud.google.com/products"},{"id":"gcp-managed-service-for-apache-spark","provider":"gcp","name":"Managed Service for Apache Spark","category":"data-pipelines","also":["warehouse"],"tagline":"Run big data jobs on Spark without managing the machines yourself.","description":"Managed Service for Apache Spark, formerly called Dataproc, is Google's hosted version of Spark, a popular tool for crunching very large amounts of data across many computers. If you have years of customer data to combine and clean before analysis, you can run a Spark job here instead of setting up your own servers. You can run it serverless or on clusters, and it also supports Hadoop and related tools. A faster Lightning Engine speeds up the work. Pick it when your team already uses Spark or wants to move existing Spark jobs to Google Cloud.","useCases":["Run Apache Spark data processing jobs","Process data for lakehouse tables","Speed up Spark jobs with Lightning Engine"],"pricing":"Per vCPU hour plus Spark premium.","status":"ga","url":"https://cloud.google.com/products/managed-service-for-apache-spark","source":"https://cloud.google.com/products"},{"id":"gcp-dataproc-metastore","provider":"gcp","name":"Dataproc Metastore","category":"data-pipelines","tagline":"A shared catalog that keeps track of what your data tables look like.","description":"Dataproc Metastore is a managed Hive metastore, which is a catalog that records which tables exist, what columns they have and where the data lives. If several tools, such as Spark and Hive jobs, need to agree on the same tables, you keep that information in one place instead of in each tool. Google runs it, so you don't maintain the database behind it. Pick it when you already use Hive or Spark on Google Cloud and want one shared source of table information. Older versions of the service are being retired, so check which version you'd use.","useCases":["Share table definitions across Spark and Hive","Keep Hive metadata managed for you","Give Spark one place for table info"],"pricing":"Per service hour.","status":"ga","url":"https://docs.cloud.google.com/dataproc-metastore/docs/overview","source":"https://docs.cloud.google.com/dataproc-metastore/docs/overview"},{"id":"gcp-managed-service-for-apache-airflow","provider":"gcp","name":"Managed Service for Apache Airflow","category":"workflows","also":["data-pipelines"],"tagline":"Schedule and run multi-step data jobs in the right order.","description":"Managed Service for Apache Airflow, formerly called Cloud Composer, runs Airflow, a tool for scheduling tasks that depend on each other. You write each workflow as a DAG, which is simply a map of steps and the order they run in. If every morning you need to pull sales data, clean it and refresh a report, Airflow makes sure each step runs after the one before it. Pick it over a basic scheduler when you have many connected steps, such as data pipelines or machine learning workflows. Airflow 3 is available in preview.","useCases":["Schedule and monitor data pipelines","Coordinate machine learning workflow steps","Run Airflow without managing the servers"],"pricing":"Environment compute and storage.","status":"ga","url":"https://cloud.google.com/products/managed-service-for-apache-airflow","source":"https://cloud.google.com/products"},{"id":"gcp-orchestration-pipelines","provider":"gcp","name":"Orchestration Pipelines","category":"workflows","also":["data-pipelines"],"tagline":"Define data and AI pipelines in simple files and deploy them safely.","description":"Orchestration Pipelines lets you describe a data or AI pipeline, a series of steps that move and process data, in YAML, a plain text format. Your pipelines, environments and bundles then deploy to Airflow, BigQuery or Spark. If your team keeps a test version and a live version of a data setup, you can store both in version control and push changes through CI/CD, an automated process for checking and releasing updates. Pick it when you want pipelines treated like app code, with a clear history and repeatable releases. It is still in preview, so expect it to evolve.","useCases":["Define pipelines in version-controlled YAML files","Deploy pipelines across dev and prod","Push data jobs to BigQuery or Spark"],"pricing":"See product page.","status":"preview","url":"https://docs.cloud.google.com/orchestration-pipelines/overview","source":"https://docs.cloud.google.com/orchestration-pipelines/overview"},{"id":"gcp-managed-service-for-apache-kafka","provider":"gcp","name":"Managed Service for Apache Kafka","category":"queues","also":["data-pipelines"],"tagline":"Run Apache Kafka for streams of app events, with Google handling setup.","description":"Managed Service for Apache Kafka runs Kafka, a popular open source system for passing large streams of events between parts of your software. If your store produces orders, payments and shipping updates that several systems need to react to, Kafka carries them between those systems. Google runs the clusters, which autoscale, and includes Connect, a way to link Kafka to other systems. Pick it over Pub/Sub when you already use Kafka or want to move an existing Kafka-based app to Google Cloud without rewriting it.","useCases":["Stream events between your services","Move existing Kafka setups to Google","Run Kafka clusters that scale automatically"],"pricing":"Per vCPU, memory and storage hour.","status":"ga","url":"https://cloud.google.com/products/managed-service-for-apache-kafka","source":"https://cloud.google.com/products"},{"id":"gcp-dataform","provider":"gcp","name":"Dataform","category":"data-pipelines","tagline":"Turn raw data into clean tables in BigQuery using just SQL.","description":"Dataform helps you reshape data that already sits in BigQuery, Google's data warehouse, using SQL, the standard language for asking questions of data. You write transformations as SQLX files, keep them in version control and schedule them to run. If your raw sales data needs to become a tidy table for reporting, you'd build that step here. It also supports assertions, which are checks that flag bad data, like missing IDs. Pick it when your data is already in BigQuery and you want a simple tool for it. Dataform itself is free, though BigQuery costs apply.","useCases":["Model data transformations in SQL","Test data for quality issues","Schedule and version SQL pipelines"],"pricing":"Free; BigQuery costs apply.","status":"ga","url":"https://cloud.google.com/dataform","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-data-fusion","provider":"gcp","name":"Cloud Data Fusion","category":"data-pipelines","tagline":"Build data pipelines by dragging and dropping, with no code needed.","description":"Cloud Data Fusion is a visual builder for ETL, which means pulling data from one place, cleaning it up and loading it somewhere else. It is based on CDAP, an open source data tool. If you want to copy data from your CRM and a few spreadsheets into a warehouse, you drag and drop steps on a canvas and pick from over 100 connectors. Pick it when people on your team aren't comfortable writing code, or when you need to bring in data from systems outside Google Cloud. Pricing is per instance hour, depending on the edition.","useCases":["Build data pipelines by dragging blocks","Connect over 100 data sources","Move data between on-site and cloud"],"pricing":"Per instance hour by edition.","status":"ga","url":"https://cloud.google.com/data-fusion","source":"https://cloud.google.com/products"},{"id":"gcp-datastream","provider":"gcp","name":"Datastream","category":"data-pipelines","tagline":"Keep a copy of your database up to date in BigQuery as changes happen.","description":"Datastream copies changes from your database to another place as they happen, a technique called change data capture. It reads from MySQL, PostgreSQL, Oracle and SQL Server, and sends the changes to BigQuery or Cloud Storage. If your app runs on PostgreSQL and you want to analyze orders without querying the live database, Datastream keeps a fresh copy in BigQuery. It is serverless, so there are no machines for you to manage, and you pay per GB processed. Pick it over manual exports when you want your reporting data to stay current on its own.","useCases":["Copy database changes to BigQuery","Keep databases in sync continuously","Replicate MySQL and PostgreSQL data"],"pricing":"Per GB processed.","status":"ga","url":"https://cloud.google.com/datastream","source":"https://cloud.google.com/products"},{"id":"gcp-pubsub","provider":"gcp","name":"Pub/Sub","category":"queues","tagline":"Send messages between parts of your app, or to many listeners at once.","description":"Pub/Sub is a messaging service where one part of your system publishes a message and others receive it. If a customer places an order, you can announce it once and let billing, email and analytics each pick it up. It supports push and pull delivery, ordering, exactly-once delivery and subscriptions that write straight into BigQuery. Messages arrive at least once, so a listener may occasionally see one twice. Pick it over Kafka when you want a simple global service with little to set up. You pay per GB, and the first 10 GB each month are free.","useCases":["Send events between app components","Ingest streams of incoming data","Send one message to many subscribers"],"pricing":"Per GB. First 10 GB per month free.","status":"ga","url":"https://cloud.google.com/pubsub","source":"https://cloud.google.com/products"},{"id":"gcp-pubsub-lite","provider":"gcp","name":"Pub/Sub Lite","category":"queues","tagline":"A low-cost message queue for your app, now being shut down.","description":"Pub/Sub Lite is a budget message queue, a way for one part of your app to hand messages to another, built around a single Google Cloud zone (one data center area) with low cost in mind. If you already send a stream of events through it, like clicks or sensor readings, it keeps working for you today. It is closed to new customers, and the docs say it will be turned down in January 2027, after earlier dates. For anything new, use Managed Kafka or Pub/Sub instead.","useCases":["Keep existing Pub/Sub Lite streams running","Move legacy streams to Kafka","Switch to Managed Kafka or Pub/Sub"],"pricing":"Existing customers only.","status":"deprecated","url":"https://docs.cloud.google.com/pubsub/lite/docs","source":"https://docs.cloud.google.com/pubsub/lite/docs"},{"id":"gcp-blockchain-analytics","provider":"gcp","name":"Blockchain Analytics","category":"warehouse","also":["specialized"],"tagline":"Public blockchain data, already organized and ready to search in BigQuery.","description":"Blockchain Analytics gives you public blockchain data, the open record of crypto transactions, already indexed and sitting in BigQuery, Google's tool for asking questions of large tables of data. If you are building a product that tracks wallet activity or token trends, you can query that data with SQL, a common data-question language, instead of collecting and cleaning it yourself. Pick it when you want answers about on-chain activity without running your own blockchain infrastructure. You pay normal BigQuery query costs.","useCases":["Analyze public blockchain transactions in SQL","Research wallet activity and token trends","Query indexed chain data directly"],"pricing":"BigQuery query costs.","status":"ga","url":"https://cloud.google.com/blockchain-analytics","source":"https://cloud.google.com/products"},{"id":"gcp-cortex-framework","provider":"gcp","name":"Cortex Framework","category":"warehouse","also":["specialized"],"tagline":"Ready-made data setups for SAP, Salesforce and marketing data in BigQuery.","description":"Cortex Framework is a set of prebuilt templates that organize business data from tools like SAP, Salesforce and marketing platforms inside BigQuery, Google's data warehouse (a place to store and analyze large tables). If you want a sales or marketing dashboard but dread designing the data tables yourself, it hands you a starting structure, plus Looker blocks, which are ready-made dashboard pieces for Google's Looker reporting tool. Choose it over building from scratch when your data already lives in those systems. The framework is free; you pay for BigQuery usage.","useCases":["Start with prebuilt SAP data models","Analyze Salesforce data in BigQuery","Build marketing reports with Looker blocks"],"pricing":"Free framework; BigQuery usage.","status":"ga","url":"https://cloud.google.com/solutions/cortex","source":"https://cloud.google.com/products"},{"id":"gcp-manufacturing-data-engine","provider":"gcp","name":"Manufacturing Data Engine","category":"specialized","also":["data-pipelines","iot"],"tagline":"Pull in factory equipment data and make sense of it for analysis.","description":"Manufacturing Data Engine collects data from machines and equipment on a factory floor and adds context, so raw numbers become something you can analyze. If you are building a product for plant operators, such as a view of how equipment is performing, you can send that data into BigQuery, Google's tool for analyzing large datasets. It is aimed at industrial settings, so it fits better than a general-purpose database when your data starts on factory equipment. Pricing is handled by contacting Google's sales team.","useCases":["Bring factory equipment data into analytics","Connect plant machines to BigQuery","Add context to factory sensor data"],"pricing":"Contact sales.","status":"ga","url":"https://docs.cloud.google.com/manufacturing-data-engine/docs","source":"https://docs.cloud.google.com/manufacturing-data-engine/docs"},{"id":"gcp-spanner","provider":"gcp","name":"Spanner","category":"sql","tagline":"A database that scales worldwide and keeps everyone's data in agreement.","description":"Spanner is a database that stores data in tables and can spread across many machines, even across continents, while every user still sees the same up-to-date data. If you run a payments app, a financial ledger or a game with players worldwide, and cannot afford mismatched balances or scores, it is a strong fit. It uses SQL, the standard database language, promises 99.999 percent uptime, and includes graph, vector (finding similar items for AI features) and full-text search. You pay per node or processing unit hour, with a 90-day free trial.","useCases":["Run a global transaction database","Track account balances in ledgers","Store player data for online games"],"pricing":"Per node or processing unit hour. 90-day free trial.","status":"ga","url":"https://cloud.google.com/spanner","source":"https://cloud.google.com/products","gem":"The database Google runs its own global products on, rentable by the hour."},{"id":"gcp-spanner-omni","provider":"gcp","name":"Spanner Omni","category":"sql","tagline":"Spanner, the global database, that you can run outside Google Cloud.","description":"Spanner Omni is a downloadable edition of Spanner, Google's globally scalable database, that you run in your own environment instead of on Google Cloud. If your customers need data kept in their own data center, or you work at the edge (close to where data is created), it lets you use the same database there. Pick it when you want Spanner but cannot move everything to Google's cloud. It is currently in preview, meaning it is still early access. See the product page for pricing.","useCases":["Run Spanner in your own data center","Run Spanner at the edge or on-site","Set up hybrid Spanner across environments"],"pricing":"See product page.","status":"preview","url":"https://docs.cloud.google.com/spanner-omni","source":"https://docs.cloud.google.com/spanner-omni/docs"},{"id":"gcp-cloud-sql","provider":"gcp","name":"Cloud SQL","category":"sql","tagline":"Managed MySQL, PostgreSQL and SQL Server, so Google handles the upkeep.","description":"Cloud SQL runs popular databases, MySQL, PostgreSQL and SQL Server, for you, so you do not have to set up or look after the machines yourself. If you are building a web app that stores users, orders or posts, it is a straightforward home for that data. It offers automated backups, high availability (staying online if a machine fails) and replicas (extra copies of your data), with Enterprise Plus editions. Pick it when you want a familiar database or an easy move for an existing app. You pay by the hour for processing power, memory and storage.","useCases":["Run web app databases with automatic backups","Move existing MySQL or PostgreSQL databases over","Keep databases running with high availability"],"pricing":"Per vCPU, memory and storage hour.","status":"ga","url":"https://cloud.google.com/sql","source":"https://cloud.google.com/products"},{"id":"gcp-alloydb","provider":"gcp","name":"AlloyDB for PostgreSQL","category":"sql","also":["vector"],"tagline":"A PostgreSQL-compatible database built for busy, demanding apps.","description":"AlloyDB for PostgreSQL is Google's database that works with Postgres, a very popular open-source database, but is built for heavier workloads. If your app handles lots of orders and also needs quick reports on them, or you are adding AI features that find similar items (vector search), it covers both in one place. It includes a columnar engine, which speeds up analysis questions, ScaNN vector search, and 99.99 percent availability. Pick it over Cloud SQL when plain Postgres starts to feel slow. You pay by the hour for processing power, memory and storage.","useCases":["Run one database for transactions and analytics","Build AI apps on PostgreSQL","Power demanding apps with very high uptime"],"pricing":"Per vCPU, memory and storage hour.","status":"ga","url":"https://cloud.google.com/products/alloydb","source":"https://cloud.google.com/products"},{"id":"gcp-alloydb-omni","provider":"gcp","name":"AlloyDB Omni","category":"sql","tagline":"AlloyDB you can run on any cloud, your own servers or your laptop.","description":"AlloyDB Omni is a downloadable version of AlloyDB, Google's database that works with Postgres (a popular open-source database), and it runs on any cloud, in your own data center or on a laptop. If you want to build and test your app against it on your own computer, or your customers need it inside their setup, you can do that. Choose it over the hosted AlloyDB when you cannot or do not want to run on Google Cloud. It is free for development use, and production use needs a license.","useCases":["Run AlloyDB on your own servers","Test and develop on a laptop","Use PostgreSQL across clouds and on-site"],"pricing":"Free dev use; production license.","status":"ga","url":"https://cloud.google.com/alloydb/omni","source":"https://cloud.google.com/products"},{"id":"gcp-bigtable","provider":"gcp","name":"Bigtable","category":"nosql","tagline":"A huge, fast database for nonstop streams of data like sensor readings.","description":"Bigtable is a database built to hold enormous amounts of data, up to petabytes (millions of gigabytes), and respond in a few thousandths of a second. If your product collects a constant flow of events, like readings from connected devices or ad clicks, it is made for that kind of load. It is a wide-column NoSQL database, meaning data is not kept in the usual linked tables, and it supports SQL and the HBase API (compatible with an older big-data tool). Pick it for time series, ad tech and IoT data at large scale. You pay per node hour plus storage.","useCases":["Store time series readings at huge scale","Handle ad tech data with fast lookups","Collect sensor data from IoT devices"],"pricing":"Per node hour plus storage.","status":"ga","url":"https://cloud.google.com/bigtable","source":"https://cloud.google.com/products"},{"id":"gcp-firestore","provider":"gcp","name":"Firestore","category":"nosql","also":["realtime-db"],"tagline":"A flexible app database that keeps screens updated in real time.","description":"Firestore is a database that stores your data as flexible documents, like small records, instead of rigid tables, and you do not manage any servers. If you are building a mobile app or a chat feature where changes should show up instantly on every screen, its live listeners and offline support handle that. It comes with SDKs, ready-made code kits for your app. Pick it when you want to build fast with real-time sync. The Enterprise edition adds MongoDB compatibility. You pay per operation and storage, with a daily free quota.","useCases":["Store data for mobile apps","Sync live updates to users' screens","Keep app data working offline"],"pricing":"Per operation and storage. Daily free quota.","status":"ga","url":"https://cloud.google.com/firestore","source":"https://cloud.google.com/products"},{"id":"gcp-firestore-mongodb-compatibility","provider":"gcp","name":"Firestore with MongoDB compatibility","category":"nosql","tagline":"Run your MongoDB app on Firestore without running MongoDB servers.","description":"This lets apps written for MongoDB, a widely used document database, run on Firestore Enterprise edition using the same MongoDB drivers and tools. If you built your product on MongoDB and are tired of looking after your own database servers, you can point your app at Firestore instead of rewriting it. Pick it when you want to move off self-managed MongoDB but keep your existing code largely as it is. You pay per operation and storage.","useCases":["Run MongoDB apps without MongoDB servers","Move off self-managed MongoDB servers","Use existing MongoDB drivers and tools"],"pricing":"Per operation and storage.","status":"ga","url":"https://docs.cloud.google.com/firestore/mongodb-compatibility/docs/overview","source":"https://docs.cloud.google.com/firestore/mongodb-compatibility/docs/overview"},{"id":"gcp-datastore","provider":"gcp","name":"Firestore in Datastore mode","category":"nosql","tagline":"A flexible database for older App Engine apps, built on Firestore.","description":"Firestore in Datastore mode stores your data as flexible records instead of rigid tables, and it speaks the older Datastore language that many apps were written for. If you already run an App Engine app (Google's service for hosting web apps) or have existing App Engine data, this is where that data naturally lives. It is built to grow with your traffic without you managing servers. Pick it when you need to keep existing Datastore code or data working. You pay per operation and for storage.","useCases":["Keep App Engine apps on their data","Store records for App Engine style apps","Scale key-based data for legacy apps"],"pricing":"Per operation and storage.","status":"ga","url":"https://cloud.google.com/datastore","source":"https://cloud.google.com/products"},{"id":"gcp-memorystore-redis","provider":"gcp","name":"Memorystore for Redis","category":"kv-cache","tagline":"A ready-made speed layer that keeps your app's busiest data in memory.","description":"Memorystore for Redis is a managed version of Redis, a popular tool that holds data in memory so your app can fetch it almost instantly. If your app keeps asking the database the same questions, or you need to remember who is logged in, a cache or session store like this takes the load off. It comes in Basic and Standard tiers, and Standard adds high availability, meaning a backup takes over if the main one fails. You pay per GB of memory per hour.","useCases":["Cache frequently used data for faster pages","Store user sessions in memory","Run managed Redis with high availability"],"pricing":"Per GB hour.","status":"ga","url":"https://cloud.google.com/memorystore","source":"https://cloud.google.com/products"},{"id":"gcp-memorystore-redis-cluster","provider":"gcp","name":"Memorystore for Redis Cluster","category":"kv-cache","tagline":"Redis spread across many machines, for caches too big for just one.","description":"Memorystore for Redis Cluster is a managed Redis setup that splits your data across several machines, so it can hold far more than a single one could. If you run a game with a live leaderboard or a very large cache, this is the version built for that size. It scales to terabytes and comes with a 99.99 percent uptime promise (called an SLA). Choose it over plain Memorystore for Redis when one machine's memory is no longer enough. You pay per node hour.","useCases":["Cache very large datasets across shards","Power leaderboards and rankings","Scale caches to terabytes"],"pricing":"Per node hour.","status":"ga","url":"https://cloud.google.com/memorystore/docs/cluster","source":"https://cloud.google.com/products"},{"id":"gcp-memorystore-valkey","provider":"gcp","name":"Memorystore for Valkey","category":"kv-cache","tagline":"Managed Valkey, a free open source stand-in for Redis.","description":"Memorystore for Valkey is a managed version of Valkey, an open source data store that works with Redis tools and code. It keeps data in memory so your app can read it very quickly. If your app needs a fast cache or a real-time store for things like live counters, you can use it much as you would Redis. It supports Valkey 7.2 and 8 in clustered mode, which spreads data over several machines. Pick it if you prefer an open source option. You pay per node hour.","useCases":["Cache data with open source Redis-compatible software","Run real-time data stores","Cluster Valkey across multiple nodes"],"pricing":"Per node hour.","status":"ga","url":"https://cloud.google.com/memorystore/docs/valkey","source":"https://cloud.google.com/products"},{"id":"gcp-memorystore-memcached","provider":"gcp","name":"Memorystore for Memcached","category":"kv-cache","tagline":"A simple managed memory cache to make your web pages load faster.","description":"Memorystore for Memcached is a managed version of Memcached, a basic tool that keeps recent results in memory so they do not need to be rebuilt. If your website runs the same slow database query for every visitor, you can save the answer here and serve it in a flash. It is a plain, distributed cache without extras, which keeps it easy to understand. Pick it when you only need page or query result caching, and choose Redis if you need more. You pay per GB hour.","useCases":["Cache web pages for faster loading","Store query results to reuse","Run a simple distributed memory cache"],"pricing":"Per GB hour.","status":"ga","url":"https://cloud.google.com/memorystore/docs/memcached","source":"https://cloud.google.com/products"},{"id":"gcp-database-center","provider":"gcp","name":"Database Center","category":"management","also":["sql"],"tagline":"One dashboard showing the health and security of all your databases.","description":"Database Center gives you a single view of how your databases are doing across Google Cloud. It flags issues across services like Cloud SQL, AlloyDB and Spanner, which are Google's managed database products. If you have several databases and no time to check each one, this shows where something looks unhealthy or insecure. It is useful for keeping an eye on your overall security setup as your product grows. It is free, so there is little reason not to look at it.","useCases":["Monitor health of all databases","Spot security issues across databases","View your database fleet in one place"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/database-center/docs/overview","source":"https://docs.cloud.google.com/database-center/docs/overview"},{"id":"gcp-cloud-storage","provider":"gcp","name":"Cloud Storage","category":"object-storage","tagline":"Store files, images and backups, paying less for files you rarely open.","description":"Cloud Storage is where you keep files of any kind, such as photos, videos, data exports and backups. If your app lets people upload pictures, this is the usual place to put them. You choose a storage class (Standard, Nearline, Coldline or Archive) depending on how often you need the files, and rarely used data costs less to keep. Autoclass can pick the class for you, and dual-region options keep copies in two places. You pay per GB each month plus operations, and the first 5 GB are free.","useCases":["Host photos, videos and downloads","Build data lakes for analytics","Keep backups in low-cost archive classes"],"pricing":"Per GB-month by class plus ops. Free 5 GB.","status":"ga","url":"https://cloud.google.com/storage","source":"https://cloud.google.com/products"},{"id":"gcp-persistent-disk","provider":"gcp","name":"Persistent Disk","category":"block-storage","tagline":"Hard drive space for your Google Cloud servers that lives on its own.","description":"Persistent Disk is the everyday hard drive for virtual servers on Compute Engine, Google's service for renting servers. If you spin up a server to run your app, its main disk, where the operating system lives, is usually one of these. It is attached over the network, so it is separate from the server itself. You can pick Standard, balanced or SSD types depending on whether you want low cost or more speed. Choose it as the sensible default for general disk needs. You pay per GB each month.","useCases":["Provide boot disks for virtual machines","Add general-purpose storage to servers","Choose standard, balanced or SSD disks"],"pricing":"Per GB-month.","status":"ga","url":"https://cloud.google.com/persistent-disk","source":"https://cloud.google.com/products"},{"id":"gcp-hyperdisk","provider":"gcp","name":"Hyperdisk","category":"block-storage","tagline":"Fast server disks where you set speed separately from size.","description":"Hyperdisk is high-performance disk storage for Google Cloud servers where you choose the speed and the size independently. Speed is measured in IOPS (reads and writes per second) and throughput (how much data moves per second). If you run a busy database or feed large amounts of data to an AI workload, you can pay for exactly the speed you need instead of buying a bigger disk. Pick it over Persistent Disk when performance matters most. You pay per GB plus the IOPS you provision.","useCases":["Run databases that need fast disks","Set speed and IOPS independently","Store data for AI workloads"],"pricing":"Per GB plus provisioned IOPS.","status":"ga","url":"https://cloud.google.com/compute/docs/disks/hyperdisks","source":"https://cloud.google.com/products"},{"id":"gcp-local-ssd","provider":"gcp","name":"Local SSD","category":"block-storage","tagline":"Very fast temporary disk on the server itself, for scratch work.","description":"Local SSD is a fast solid state drive that sits directly on the physical machine running your server, so reads and writes have very little delay. If your app needs a quick place for temporary files, such as scratch space for processing or a speedy cache, this fits well. Because it is ephemeral, meaning temporary, you should not keep anything on it that you cannot recreate. Pick it when raw speed matters more than keeping data safe long term. You pay per GB each month.","useCases":["Provide fast scratch space for processing","Speed up temporary caches on servers","Hold temporary data that can be lost"],"pricing":"Per GB-month.","status":"ga","url":"https://cloud.google.com/products/local-ssd","source":"https://cloud.google.com/products"},{"id":"gcp-filestore","provider":"gcp","name":"Filestore","category":"block-storage","tagline":"A shared network folder that many servers can use at once.","description":"Filestore is a managed shared file system using NFS, a standard way for several machines to read and write the same folders. If you have more than one server that needs the same files, such as uploaded content or shared settings, this gives them one common place. It also works as storage volumes for GKE, Google's service for running app containers. It comes in zonal, regional and enterprise tiers, with more protection at the higher levels. Pick it when you need a real shared folder. You pay per GB each month.","useCases":["Share files across multiple servers","Give GKE containers shared file storage","Offer NFS file storage with regional options"],"pricing":"Per GB-month.","status":"ga","url":"https://cloud.google.com/filestore","source":"https://cloud.google.com/products"},{"id":"gcp-managed-lustre","provider":"gcp","name":"Google Cloud Managed Lustre","category":"block-storage","also":["gpu"],"tagline":"Very fast shared file storage for AI training and heavy computing.","description":"Google Cloud Managed Lustre is a managed file system built to let many machines read huge amounts of data at the same time. Lustre is a well known parallel file system, meaning it spreads files across many drives to move data quickly. If you are training an AI model and your graphics chips sit idle waiting for data, this keeps them fed. It also suits HPC, which means high-performance computing such as large simulations. Pick it over Filestore when raw throughput is the goal. You pay per GiB each month by tier.","useCases":["Speed up AI model training data access","Run high-performance computing jobs","Serve large parallel file workloads"],"pricing":"Per GiB-month by tier.","status":"ga","url":"https://cloud.google.com/products/managed-lustre","source":"https://cloud.google.com/products"},{"id":"gcp-netapp-volumes","provider":"gcp","name":"Google Cloud NetApp Volumes","category":"block-storage","tagline":"Shared network drives for your files, run for you by Google Cloud.","description":"Google Cloud NetApp Volumes gives you shared folders that many servers and people can open at once, using the NFS and SMB file-sharing standards. If your team has a business app that expects a regular network drive, or you are moving a workload like SAP or VMware (software many large companies run), you can point it here instead of building your own. It keeps snapshots (point-in-time copies you can roll back to) and can replicate data to another location. Pick it over plain file or object storage when your software needs a real shared drive. You pay per GiB, based on the service level you choose.","useCases":["Host enterprise file shares with snapshots","Back SAP and VMware datastores","Replicate file data to other sites"],"pricing":"Per GiB by service level.","status":"ga","url":"https://cloud.google.com/netapp-volumes","source":"https://cloud.google.com/products"},{"id":"gcp-backup-and-dr","provider":"gcp","name":"Backup and DR Service","category":"block-storage","tagline":"Back up your servers, databases and files, and recover fast after a disaster.","description":"Backup and DR Service makes protected copies of your virtual machines, databases and file systems so you can restore them when something goes wrong. If ransomware locks your data or a region goes down, you can bring your app back from a copy the attacker cannot change. Those copies live in immutable vaults, which means they cannot be edited or deleted for a set period. Choose it over doing your own copies when you want one managed place to handle backups and recovery. You pay per GB protected.","useCases":["Recover from ransomware with immutable backups","Protect VMs, databases and file systems","Plan disaster recovery for business systems"],"pricing":"Per GB protected.","status":"ga","url":"https://cloud.google.com/backup-disaster-recovery","source":"https://cloud.google.com/products"},{"id":"gcp-backup-for-gke","provider":"gcp","name":"Backup for GKE","category":"block-storage","also":["containers"],"tagline":"Save and restore your apps running on Google Kubernetes Engine.","description":"Backup for GKE saves the setup and stored data of apps running on GKE, Google's managed Kubernetes (a system for running many containers, the packaged pieces of an app). If a cluster breaks or someone deletes the wrong thing, you can restore the app instead of rebuilding it by hand. It also helps when you want to move a workload to a new cluster. Pick it over general backup tools when your app lives in GKE and you want both its settings and its volumes saved together. You pay per pod protected, plus storage.","useCases":["Back up Kubernetes cluster settings and volumes","Restore GKE clusters after failure","Move workloads to another cluster"],"pricing":"Per pod protected plus storage.","status":"ga","url":"https://docs.cloud.google.com/kubernetes-engine/docs/add-on/backup-for-gke","source":"https://docs.cloud.google.com/kubernetes-engine/docs/add-on/backup-for-gke"},{"id":"gcp-storage-transfer-service","provider":"gcp","name":"Storage Transfer Service","category":"migration","also":["object-storage"],"tagline":"Copy large amounts of data into Google Cloud on a schedule.","description":"Storage Transfer Service moves files into Cloud Storage, Google's file and object storage, from other clouds or from your own servers. If your images and backups sit in Amazon S3 and you want them in Google Cloud, you set up a transfer and it does the copying for you, once or on a repeating schedule. It beats writing your own copy scripts because Google runs and manages the job. It is also handy for moving old data into archive storage. Some transfer modes are charged per GB moved.","useCases":["Copy data from Amazon S3 into Cloud Storage","Move files from other clouds on a schedule","Move archives into Cloud Storage in bulk"],"pricing":"Per GB transferred for some modes.","status":"ga","url":"https://cloud.google.com/storage-transfer-service","source":"https://cloud.google.com/products"},{"id":"gcp-transfer-appliance","provider":"gcp","name":"Transfer Appliance","category":"migration","tagline":"Get huge amounts of data to Google Cloud by mailing a physical device.","description":"Transfer Appliance is a physical storage device that Google sends to you so you can load your data onto it and ship it back. If you have petabytes (thousands of terabytes) of data, uploading over the internet could take far too long, and this gets around that. A typical case is seeding a data lake, a big pool of raw data you plan to analyze, without waiting on your connection. Pick it when your data is too large for online transfer to be practical. You pay a rental fee for each appliance.","useCases":["Ship terabytes or petabytes on physical devices","Migrate large datasets offline by shipping","Seed a data lake with offline transfer"],"pricing":"Rental fee per appliance.","status":"ga","url":"https://cloud.google.com/transfer-appliance/docs/4.0/overview","source":"https://cloud.google.com/products"},{"id":"gcp-data-transfer-essentials","provider":"gcp","name":"Data Transfer Essentials","category":"migration","also":["networking"],"tagline":"Lower-cost ways to move your data between clouds and into Google Cloud.","description":"Data Transfer Essentials is a set of tools for moving data into Google Cloud, with low-cost paths for multicloud setups. If your app or data lives partly on another cloud and you need to shuffle information between them, this helps keep the bill down. Use it when you regularly move data across clouds and the transfer cost is what worries you. For one-off or very large moves, other transfer tools may fit better. Pricing is not spelled out here, so check the product page for current details.","useCases":["Move data between clouds at lower cost","Set up multicloud data transfer paths","Transfer data to Google Cloud more cheaply"],"pricing":"See product page.","status":"ga","url":"https://docs.cloud.google.com/data-transfer-essentials/docs","source":"https://docs.cloud.google.com/data-transfer-essentials/docs"},{"id":"gcp-workflows","provider":"gcp","name":"Workflows","category":"workflows","tagline":"Chain web services and Google tools into one automatic step-by-step process.","description":"Workflows lets you describe a series of steps in a simple YAML or JSON file, and Google runs them in order for you with no server to manage. If a new order should check inventory, charge a card, then send an email, you can wire those calls together here. It handles retries when a step fails and can wait for a callback, such as a person approving a request. Pick it over writing your own glue code when you want the process to be visible and reliable. You pay per step, and the first 5,000 internal steps are free.","useCases":["Connect several web services in order","Retry failed steps automatically","Run approval steps before continuing"],"pricing":"Per step. First 5,000 internal steps free.","status":"ga","url":"https://cloud.google.com/workflows","source":"https://cloud.google.com/products"},{"id":"gcp-eventarc","provider":"gcp","name":"Eventarc","category":"queues","also":["serverless"],"tagline":"Send a signal to your app when something happens in Google Cloud or elsewhere.","description":"Eventarc listens for things that happen, like a file landing in storage or a change in your Google Cloud account, and passes a message about it to the right place. If you want your app to react whenever a user uploads a document, you can use Eventarc to start your code at that moment. It delivers events in the CloudEvents format (a common standard for describing events) to Cloud Run, GKE and Workflows. It also works with SaaS and your own custom sources. Pick it when you want your app to respond to events instead of checking constantly. You pay per event.","useCases":["Trigger apps when events happen","Send Google Cloud audit logs to Cloud Run","Route events from SaaS tools"],"pricing":"Per event.","status":"ga","url":"https://cloud.google.com/eventarc/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-tasks","provider":"gcp","name":"Cloud Tasks","category":"queues","tagline":"A to-do queue for your app that spreads out jobs and retries failures.","description":"Cloud Tasks is a managed queue where your app drops jobs to be done later, and Google delivers them one by one to your web address or App Engine app. If signing up a user should trigger a slow welcome email, you can queue it so the user does not wait. You can cap how fast tasks go out, which protects other services from being flooded, and failed tasks are retried for you. Pick it when you need control over pacing and retries. You pay per million operations, and the first million are free.","useCases":["Queue slow work to run later","Send calls at a controlled rate","Retry tasks that fail"],"pricing":"Per million operations. First million free.","status":"ga","url":"https://cloud.google.com/tasks/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-scheduler","provider":"gcp","name":"Cloud Scheduler","category":"workflows","tagline":"Run a task on a timer, like a managed alarm clock for your app.","description":"Cloud Scheduler is a managed cron service, meaning it triggers things at times you set. If you want a report built every night or stale data cleaned up every Sunday, you tell it when and what to call. It can hit a web address, send a Pub/Sub message (Google's messaging service) or trigger an App Engine app. Pick it over running your own timer on a server, since there is nothing to keep alive. You pay per job per month, and the first 3 jobs are free.","useCases":["Run jobs on a set schedule","Send nightly messages through Pub/Sub","Call web endpoints on a timer"],"pricing":"Per job per month; 3 free.","status":"ga","url":"https://cloud.google.com/scheduler/docs","source":"https://cloud.google.com/products"},{"id":"gcp-application-integration","provider":"gcp","name":"Application Integration","category":"workflows","tagline":"Connect your business apps with drag-and-drop flows instead of code.","description":"Application Integration is a low-code tool where you build automations by arranging steps on a visual canvas. If a new deal in Salesforce should start an approval process and then update another system, you can design that flow without writing much code. It has triggers to start flows and more than 100 connectors, which are ready-made links to other apps. Choose it over custom scripts when non-engineers need to read or change the process. You pay per execution, meaning each time a flow runs.","useCases":["Build SaaS integrations visually","Set up approval flows with Salesforce data","Connect over 100 apps with visual flows"],"pricing":"Per execution.","status":"ga","url":"https://cloud.google.com/application-integration","source":"https://cloud.google.com/products"},{"id":"gcp-integration-connectors","provider":"gcp","name":"Integration Connectors","category":"devtools","tagline":"Ready-made links to apps like Salesforce, SAP and Jira.","description":"Integration Connectors gives you prebuilt links to business tools and databases, including Salesforce, SAP and Jira, so your app can read and write their data without custom code. If you want a workflow to pull customer records out of SAP, a connector handles the login and the details of talking to it. You can use it from Google Cloud workflows and integrations. Pick it over building your own link when the app is a common one and you want to save time. You pay per connection node, which is one running connection.","useCases":["Read data from Salesforce or Jira","Read SAP data inside workflows","Reach enterprise databases from your app"],"pricing":"Per connection node.","status":"ga","url":"https://cloud.google.com/integration-connectors/docs","source":"https://cloud.google.com/products"},{"id":"gcp-apigee","provider":"gcp","name":"Apigee","category":"waf","also":["devtools"],"tagline":"Run, protect and sell access to your APIs from one place.","description":"Apigee is Google Cloud's full toolkit for managing APIs, the doorways that let other software talk to your product. If you offer a public API to customers or share data with partners, Apigee sits in front of it to control who gets in, track how it is used and block abuse. It also supports charging for API access, and it can front AI services you expose. Pick it over a simpler gateway when you need analytics, security and monetization together. You can try it free for evaluation, then pay by subscription or as you go.","useCases":["Publish APIs for outside developers","Charge for API access","Run proxies for AI APIs"],"pricing":"Evaluation free; subscription or pay-as-you-go.","status":"ga","url":"https://cloud.google.com/apigee","source":"https://cloud.google.com/products"},{"id":"gcp-api-gateway","provider":"gcp","name":"API Gateway","category":"devtools","tagline":"A simple front door that adds sign-in and usage limits to your API.","description":"API Gateway is a lightweight service that stands in front of your backend and decides who may call it and how often. If you built a small API on Cloud Run or Cloud Functions (Google's services for running your code without managing servers), you can use it to require sign-in and set usage limits without writing that logic yourself. Choose it over Apigee when your needs are modest and you want something quick to set up. You pay per million calls, so a small project stays cheap.","useCases":["Add login checks to serverless APIs","Set usage quotas for Cloud Run","Put one endpoint in front of Cloud Functions"],"pricing":"Per million calls.","status":"ga","url":"https://cloud.google.com/api-gateway/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-endpoints","provider":"gcp","name":"Cloud Endpoints","category":"devtools","tagline":"An older way to add usage limits and checks to your API.","description":"Cloud Endpoints is an API management tool that works from an OpenAPI file, a standard document describing what your API offers. If you already run your API on GKE (Google's managed Kubernetes, a system for running containers) or on Compute Engine virtual servers, it puts a proxy in front to apply quotas and checks. It is described as the older option, so for a new serverless project API Gateway is usually the more natural pick. You pay per million requests.","useCases":["Maintain older API proxies","Set quotas from OpenAPI definitions","Proxy APIs running on GKE"],"pricing":"Per million requests.","status":"ga","url":"https://cloud.google.com/endpoints/docs","source":"https://cloud.google.com/products"},{"id":"gcp-vpc","provider":"gcp","name":"Virtual Private Cloud (VPC)","category":"networking","tagline":"Your private network inside Google Cloud, shared across all regions.","description":"Virtual Private Cloud, or VPC, is the private network your Google Cloud resources live on. If you run servers or a GKE cluster (Google's managed Kubernetes), a VPC decides which of them can talk to each other and to the internet, using firewall rules and routes. One network spans every region, and Shared VPC lets several projects use it together. Nearly every other Google Cloud service builds on it. The network itself is free, but traffic leaving it, NAT and addresses are charged.","useCases":["Build private networks for VMs","Share one network across projects","Create hubs for hybrid networks"],"pricing":"Network itself is free. Egress, NAT and addresses are charged.","status":"ga","url":"https://cloud.google.com/vpc","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-load-balancing","provider":"gcp","name":"Cloud Load Balancing","category":"load-balancing","tagline":"Spread visitors across your servers behind one worldwide address.","description":"Cloud Load Balancing splits incoming traffic across several copies of your app so no single one gets overwhelmed. If your website suddenly gets a burst of visitors, it scales up on its own without needing to be warmed up first. It works with GKE, Cloud Run and internal services, and connects to Cloud CDN and Cloud Armor (Google's attack protection). Its big draw is a single global address that serves users everywhere. You pay per rule per hour plus per GB of data processed.","useCases":["Route web traffic through one global IP","Spread traffic across GKE and Cloud Run","Balance traffic between internal services"],"pricing":"Per forwarding rule hour plus per GB processed.","status":"ga","url":"https://cloud.google.com/load-balancing","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-cdn","provider":"gcp","name":"Cloud CDN","category":"cdn","tagline":"Make your site load faster by keeping copies close to visitors.","description":"Cloud CDN keeps copies of your images, scripts and pages on Google's servers around the world, so visitors get them from somewhere nearby. If your web app serves the same files to many people, this makes it feel faster and takes load off your own servers. It works with content from Compute Engine, Cloud Storage and outside sources, and you turn it on with a checkbox on a load balancer. Pick it when you already use Cloud Load Balancing. You pay for data sent out, cache fills and requests.","useCases":["Serve static assets faster worldwide","Cache web app content at the edge","Speed up file downloads"],"pricing":"Per GB egress plus cache fill and request charges.","status":"ga","url":"https://cloud.google.com/cdn","source":"https://cloud.google.com/products"},{"id":"gcp-media-cdn","provider":"gcp","name":"Media CDN","category":"cdn","also":["video"],"tagline":"Deliver video and very large downloads quickly to viewers worldwide.","description":"Media CDN is a delivery network built for streaming video and big files. If your product serves movies, live streams or game and software downloads, it keeps copies close to viewers so playback starts fast. It uses the same edge caching setup as YouTube, and adds signed requests (links that only work for approved users), origin shielding to protect your storage, and tuning for large files. Choose it over Cloud CDN when video or huge files are your main traffic. You pay per GB by region plus requests.","useCases":["Stream video on demand","Deliver live streams to viewers","Distribute large games and software"],"pricing":"Per GB egress by region plus request charges.","status":"ga","url":"https://cloud.google.com/media-cdn","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-dns","provider":"gcp","name":"Cloud DNS","category":"dns","tagline":"Point your domain name to your app, on Google's global network.","description":"Cloud DNS is the phone book that turns your domain name, like yourapp.com, into the address of your server. If you launch a site, you add records here so visitors reach the right place. It also supports private zones that only work inside your own network, plus routing by location or weight to split traffic. Google promises 100 percent availability for it, and it supports DNSSEC, which guards against forged answers. You pay per zone each month plus per million lookups.","useCases":["Host public domain records","Resolve names privately inside VPCs","Route traffic by region or weight"],"pricing":"Per zone per month plus per million queries.","status":"ga","url":"https://cloud.google.com/dns","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-domains","provider":"gcp","name":"Cloud Domains","category":"domains","tagline":"Buy and manage your domain names right inside Google Cloud.","description":"Cloud Domains lets you register a domain name, like yourapp.com, from the Google Cloud console and set it up with Cloud DNS. If you want your domain and your hosting billed in one place, this saves you juggling a separate registrar. Renewals show up on your Google Cloud bill. It is not the same as the old Google Domains retail registrar, which Squarespace took over. Pricing is per domain per year and depends on the ending, such as .com or .io.","useCases":["Register domains from the console","Link new domains to Cloud DNS","Renew domains billed with GCP"],"pricing":"Per domain per year, by TLD.","status":"ga","url":"https://cloud.google.com/domains/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-nat","provider":"gcp","name":"Cloud NAT","category":"networking","tagline":"Let private servers reach the internet without being exposed to it.","description":"Cloud NAT lets servers and containers that have no public address still make outgoing connections, for example to download updates or call another company's API. If your private GKE cluster (Google's managed Kubernetes) needs to reach the outside, this does it with no extra proxy servers to run. It also gives you fixed outgoing addresses, handy when a partner wants to allowlist your traffic. It grows with your traffic. You pay per gateway hour plus per GB processed.","useCases":["Give private servers internet access","Use fixed IPs for allowlists","Let private GKE pods reach out"],"pricing":"Per gateway hour plus per GB processed.","status":"ga","url":"https://cloud.google.com/nat","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-router","provider":"gcp","name":"Cloud Router","category":"networking","tagline":"Automatically share network routes between Google Cloud and your offices.","description":"Cloud Router lets your Google Cloud network and your other networks, like an office or data center, swap route information automatically using BGP, a standard way networks tell each other where traffic should go. If you connect through a VPN or Interconnect (a dedicated physical link), it keeps routes current and can switch over if a link fails. It also powers Cloud NAT behind the scenes. Most small apps will not need it directly. There is no separate charge for it.","useCases":["Share routes between networks dynamically","Fail over VPN links automatically","Route traffic to on-prem networks"],"pricing":"No separate charge.","status":"ga","url":"https://cloud.google.com/network-connectivity/docs/router","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-vpn","provider":"gcp","name":"Cloud VPN","category":"networking","tagline":"Connect your office or another cloud to Google Cloud over an encrypted link.","description":"Cloud VPN creates encrypted tunnels over the public internet between your Google Cloud network and somewhere else, like your office or another cloud. If you need a private link to a database or system you cannot move yet, it is a quick way to get one running. The HA option comes with a 99.99 percent availability target, and it can serve as a backup for Interconnect, a dedicated physical connection. Choose it when you want a fast, lower-cost link. You pay per tunnel hour plus outgoing data.","useCases":["Connect offices to Google Cloud","Set up encrypted tunnels quickly","Back up a dedicated Interconnect link"],"pricing":"Per tunnel hour plus egress.","status":"ga","url":"https://cloud.google.com/network-connectivity/docs/vpn","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-interconnect","provider":"gcp","name":"Cloud Connectivity (Interconnect)","category":"networking","tagline":"Private, high-speed cable-style links between your network and Google.","description":"Cloud Connectivity (Interconnect) gives you a private line between your own network and Google Cloud, so your data skips the public internet. If your company runs systems in its own office or data center and moves lots of data to and from Google, this keeps that traffic fast and steady. Speeds range from 10 to 400 Gbps (very fast connections). Cross-Cloud Interconnect also links Google directly to other clouds, such as AWS or Azure. Pick it over a regular internet connection when volume is high. You pay per link hour plus data leaving Google.","useCases":["Move high-volume traffic privately","Link directly to AWS or Azure","Connect with 10 to 400 Gbps"],"pricing":"Per link hour plus egress.","status":"ga","url":"https://cloud.google.com/hybrid-connectivity","source":"https://cloud.google.com/products"},{"id":"gcp-network-connectivity-center","provider":"gcp","name":"Network Connectivity Center","category":"networking","tagline":"One central hub that links your cloud networks, offices and VPNs.","description":"Network Connectivity Center is a single hub that connects all your separate networks, so they can talk to each other through Google's own backbone (its private global network). If your company has several cloud networks plus a few branch offices, you can plug each one into the hub instead of wiring every pair together by hand. It also lets sites send data to each other through Google. Pick it when your connections are getting tangled. You pay per connected network, per hour, plus the data processed.","useCases":["Connect branch offices to the cloud","Mesh several VPC networks together","Link VPNs and SD-WAN sites centrally"],"pricing":"Per spoke hour plus data processed.","status":"ga","url":"https://cloud.google.com/network-connectivity-center","source":"https://cloud.google.com/products"},{"id":"gcp-network-intelligence-center","provider":"gcp","name":"Network Intelligence Center","category":"observability","also":["networking"],"tagline":"See your network map, test connections and find why things can't connect.","description":"Network Intelligence Center is a set of tools for seeing and checking your network on Google Cloud. If part of your app suddenly can't reach another part, you can run a connectivity test to find where the path breaks. It also draws a map of your network, tracks delays, and helps you review firewall rules (the rules that allow or block traffic) so you can clean up old ones. Most of it is free, though some tools charge per test or per log.","useCases":["Debug why connections fail","Test whether traffic can reach","Review and clean up firewall rules"],"pricing":"Mostly free; some modules billed per test or log.","status":"ga","url":"https://cloud.google.com/network-intelligence-center","source":"https://cloud.google.com/products"},{"id":"gcp-network-service-tiers","provider":"gcp","name":"Network Service Tiers","category":"networking","tagline":"Pick faster or cheaper internet delivery for your app's outgoing traffic.","description":"Network Service Tiers lets you choose how data travels from Google Cloud to your users. Premium sends it over Google's own private global network, which keeps delays low. Standard uses the regular public internet at a lower price. If your app is a live video tool or a game where speed matters, Premium is the safer pick. If you serve mostly simple pages and want to trim the bill, Standard can save money. You pay per GB of outgoing data, and Standard costs less.","useCases":["Lower egress costs with Standard tier","Speed up latency-sensitive apps","Choose premium or standard egress"],"pricing":"Per GB egress, cheaper on Standard tier.","status":"ga","url":"https://cloud.google.com/network-tiers","source":"https://cloud.google.com/products"},{"id":"gcp-private-service-connect","provider":"gcp","name":"Private Service Connect","category":"networking","tagline":"Reach Google services and partner apps privately, without the open internet.","description":"Private Service Connect gives your private network its own entry point to a Google service or a partner's software, using internal addresses only. If your app needs to call a Google API or a third-party service and you don't want that traffic crossing the public internet, you set up one of these endpoints. It avoids VPC peering (a way of joining two private networks that can get messy to manage). Pick it for cleaner, safer access. You pay per endpoint hour plus the data processed.","useCases":["Reach Google APIs privately","Use SaaS tools over private IPs","Connect without VPC peering"],"pricing":"Per endpoint hour plus data processed.","status":"ga","url":"https://cloud.google.com/private-service-connect","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-ngfw","provider":"gcp","name":"Cloud NGFW","category":"networking","also":["waf"],"tagline":"A built-in firewall that protects each server in your cloud network.","description":"Cloud NGFW is a firewall (a gatekeeper that decides which traffic is allowed) built right into your Google Cloud network. If you run several servers and want each one to accept only the traffic it needs, you can set rules per server. Higher tiers also filter by website name, country and known threats. Unlike older setups, it needs no extra box sitting in the traffic path. The Standard tier is free, while the Enterprise tier costs per endpoint hour plus data.","useCases":["Segment traffic between individual VMs","Block threats inside your VPC","Filter traffic by domain name"],"pricing":"Standard is free; Enterprise tier is per endpoint hour plus data.","status":"ga","url":"https://cloud.google.com/firewall","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-ids","provider":"gcp","name":"Cloud IDS","category":"threat-intel","also":["networking"],"tagline":"Spots attacks and malware in your cloud network traffic for you.","description":"Cloud IDS is a managed alarm system that watches a copy of your network traffic for signs of attack. If you must show an auditor that you monitor for threats, or you want to catch an intruder moving between your servers, this does the watching. It uses threat detection technology from Palo Alto Networks to find malware, spyware and exploits (tricks that abuse software flaws). Pick it when you want detection without running your own tools. You pay per endpoint hour plus per GB inspected.","useCases":["Detect malware in network traffic","Spot lateral movement between machines","Meet intrusion detection compliance rules"],"pricing":"Per endpoint hour plus per GB inspected.","status":"ga","url":"https://cloud.google.com/intrusion-detection-system","source":"https://cloud.google.com/products"},{"id":"gcp-secure-web-proxy","provider":"gcp","name":"Secure Web Proxy","category":"zero-trust","also":["networking"],"tagline":"Control which websites your cloud servers are allowed to visit.","description":"Secure Web Proxy is a managed checkpoint for outgoing web traffic from your cloud setup. If you want your servers to reach only a short list of approved sites, you can allow or block requests by web address and by who is asking. It can also inspect encrypted traffic and records every request, which helps when you need a clear audit trail. Pick it when you need tighter control over what leaves your network. You pay per gateway hour plus per GB.","useCases":["Limit which websites servers can visit","Log every outbound web request","Block access to unwanted sites"],"pricing":"Per gateway hour plus per GB.","status":"ga","url":"https://cloud.google.com/secure-web-proxy","source":"https://cloud.google.com/products"},{"id":"gcp-secure-access-connect","provider":"gcp","name":"Secure Access Connect","category":"zero-trust","also":["networking"],"tagline":"Secure links from offices and remote staff into Google Cloud.","description":"Secure Access Connect is a service for linking branch offices and remote users to Google Cloud with security checks along the way. It follows the SASE approach (bundling network access and security into one cloud service) and works with partner security tools. If your team is spread across offices and home, it can help keep access safe. It is in preview, so features may still change. Pricing is not listed here, so contact sales or check the pricing page.","useCases":["Connect branch offices to cloud securely","Link remote workers to the cloud","Plug in partner security services"],"pricing":"Contact sales; check pricing page.","status":"preview","url":"https://docs.cloud.google.com/secure-access-connect/docs","source":"https://docs.cloud.google.com/secure-access-connect/docs"},{"id":"gcp-network-security-integration","provider":"gcp","name":"Network Security Integration","category":"waf","also":["networking"],"tagline":"Plug third-party firewalls and security tools into your cloud traffic.","description":"Network Security Integration sends your cloud traffic through security products from other companies. If your team already trusts a particular firewall brand, you can route traffic through it instead of switching to Google's own tools. It can place the tool directly in the traffic path or just give it a copy to inspect, such as for intrusion detection. It builds on Google Cloud's load balancing, which spreads traffic across servers. You pay for the partner product plus Google networking.","useCases":["Route traffic through third-party firewalls","Copy traffic to intrusion detection tools","Insert partner security appliances into networks"],"pricing":"Pay for partner products and Google networking.","status":"ga","url":"https://docs.cloud.google.com/network-security-integration/docs","source":"https://docs.cloud.google.com/network-security-integration/docs"},{"id":"gcp-service-extensions","provider":"gcp","name":"Service Extensions","category":"edge-compute","also":["networking"],"tagline":"Add your own custom rules to Google's traffic and delivery layer.","description":"Service Extensions lets you run your own small pieces of logic inside Google's load balancers (which spread traffic across servers) and Cloud CDN (which delivers content from locations near users). If you want to check a login before a request reaches your app, rewrite page headers, or block bad bots, you can do it right there. It runs WebAssembly plugins (compact, fast add-on programs) or calls out to other services. Pick it when standard settings aren't flexible enough. Pricing is usage-based, so see the pricing page.","useCases":["Add custom login checks near users","Rewrite headers on incoming requests","Run bot filtering inside the CDN"],"pricing":"Usage-based; see pricing page.","status":"ga","url":"https://docs.cloud.google.com/service-extensions/docs/overview","source":"https://docs.cloud.google.com/service-extensions/docs/overview"},{"id":"gcp-service-directory","provider":"gcp","name":"Service Directory","category":"networking","tagline":"A phone book that helps your services find each other anywhere.","description":"Service Directory is a managed list of your services and where to find them. If your app has several parts, some on Google Cloud and some in your own data center or another cloud, each part can look up the others by name instead of a fixed address. It supports lookups through DNS (the internet's naming system), HTTP and gRPC (a fast way for programs to talk). Pick it when your setup spans several places and names keep getting out of sync. You pay per million API operations.","useCases":["Let services find each other","Name services across different clouds","Look up services by name"],"pricing":"Per million API operations.","status":"ga","url":"https://cloud.google.com/service-directory","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-armor","provider":"gcp","name":"Cloud Armor","category":"waf","also":["ddos"],"tagline":"Shields your web app from traffic floods and common attacks.","description":"Cloud Armor is a shield that sits in front of your Google Cloud app and filters out bad traffic before it arrives. If you run a public website or API and worry about floods of fake traffic (known as a DDoS attack) or bots hammering your sign-up page, you can turn on ready-made rules and cap how many requests one visitor can make. It works at the load balancer, the front door that spreads visitors across your servers, so it suits apps already running on Google Cloud. Pricing is per policy and rule, plus per million requests, and an Enterprise subscription is available.","useCases":["Block distributed denial of service attacks","Filter attacks on web apps","Limit how often APIs get called"],"pricing":"Per policy and rule, plus per million requests. Enterprise subscription available.","status":"ga","url":"https://cloud.google.com/security/products/armor","source":"https://cloud.google.com/products"},{"id":"gcp-vpc-service-controls","provider":"gcp","name":"VPC Service Controls","category":"networking","also":["secrets"],"tagline":"Draws a fence so your data can't be copied out of Google Cloud.","description":"VPC Service Controls draws a protective boundary around your Google Cloud projects and services such as BigQuery (a data analysis tool) and Cloud Storage (file storage). If your app holds customer records or other regulated data, you can keep that data reachable only from inside the boundary, which makes it much harder to copy elsewhere by mistake or misuse. That is different from ordinary permissions, which decide who may open data but not where it can end up. It is built to limit data exfiltration, meaning data leaking out. There is no charge for it.","useCases":["Stop sensitive data leaving your project","Protect regulated data with perimeters","Lock down BigQuery and storage access"],"pricing":"No charge.","status":"ga","url":"https://cloud.google.com/vpc-service-controls","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-number-registry","provider":"gcp","name":"Cloud Number Registry","category":"networking","also":["management"],"tagline":"Keeps track of your network address ranges and spots free ones.","description":"Cloud Number Registry is an address book for the IP addresses, the numbers that identify things on a network, used across your Google Cloud setup. If several teams in your company each create their own private networks (VPCs), it discovers the ranges in use across your organization, shows how full they are, and finds free ranges. That helps you plan ahead and avoid overlapping ranges, which cause headaches when you later connect networks together. It is in preview, meaning still being tested, and pricing is not final.","useCases":["Plan network address ranges","Spot overlapping IP address blocks","See how much address space is used"],"pricing":"Preview; pricing not final.","status":"preview","url":"https://docs.cloud.google.com/number-registry/overview","source":"https://docs.cloud.google.com/number-registry/overview"},{"id":"gcp-cloud-location-finder","provider":"gcp","name":"Cloud Location Finder","category":"management","tagline":"Helps you pick the best Google Cloud region to run your app in.","description":"Cloud Location Finder lists the regions and zones, the geographic spots where cloud data centers live, and shows how close they are to you along with carbon data. If most of your customers are in one part of the world, you can use it to shortlist nearby regions so your app feels faster, or to favor greener ones. It comes as an API (a way for software to ask questions) and as a console tool, and it is built to compare regions across providers, not only Google's. It is free.","useCases":["Pick a region for your app","Compare regions by distance and carbon","Choose lower carbon regions"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/location-finder/docs/overview","source":"https://docs.cloud.google.com/location-finder/docs/overview"},{"id":"gcp-cloud-iam","provider":"gcp","name":"Identity and Access Management (IAM)","category":"identity","tagline":"Decide who can do what in your Google Cloud account.","description":"IAM (Identity and Access Management) is the permission system that controls which people and programs can touch which parts of your Google Cloud setup. If you hire a freelancer to look at your app's logs, you can give them a role that lets them read logs and nothing else. It offers ready-made roles, custom roles, conditions, deny rules and workload identity federation, which lets outside systems prove who they are without a stored password. Every Google Cloud service relies on it, so you will use it whichever products you pick. It is free.","useCases":["Control who can access what","Give users only needed permissions","Create accounts for apps and services"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/security/products/iam","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-identity","provider":"gcp","name":"Cloud Identity","category":"identity","tagline":"Staff logins, single sign-on and device rules without Google Workspace.","description":"Cloud Identity is a service for managing your team's accounts, groups and logins. If you are a small company that runs on Google Cloud but does not use Google Workspace (Google's email and docs suite), it gives you a proper staff directory. You can require a second check at sign-in (MFA), let people log in once to reach other business apps (SSO), and manage employees' devices. Pick it when you need those staff controls without buying Workspace. There is a free edition, and Premium is priced per user per month.","useCases":["Manage staff accounts and groups","Add single sign-on to SaaS tools","Require two-step login for staff"],"pricing":"Free edition; Premium per user per month.","status":"ga","url":"https://cloud.google.com/identity","source":"https://cloud.google.com/products"},{"id":"gcp-identity-platform","provider":"gcp","name":"Identity Platform","category":"identity","tagline":"Add sign-in with Google, email, phone or SAML to your app.","description":"Identity Platform handles sign-up and login for the people who use your app, so you don't build it yourself. If you are launching a product where customers create accounts, you can offer social logins, email, phone, SAML or OIDC (two common business sign-in standards), plus a second check at sign-in (MFA). It is the enterprise tier behind Firebase Authentication with Identity Platform, which suits apps serving many separate customer groups. Pricing is per monthly active user, with a free allowance of 50,000 users on the base tier.","useCases":["Add sign-in for your app users","Support social, email and phone sign-in","Handle separate customer organizations"],"pricing":"Per monthly active user, with a free allowance of 50,000 users on the base tier.","status":"ga","url":"https://cloud.google.com/security/products/identity-platform","source":"https://cloud.google.com/products"},{"id":"gcp-managed-microsoft-ad","provider":"gcp","name":"Managed Service for Microsoft Active Directory","category":"identity","tagline":"Google runs your Microsoft Active Directory so you don't have to.","description":"Managed Service for Microsoft Active Directory is Google running Active Directory, Microsoft's system for managing logins and computers in a company, on your behalf. If you are moving an older Windows-based app to Google Cloud and it expects Active Directory to be there, this saves you from setting it up yourself. Google runs the domain controllers (the servers behind it), keeps them available, applies updates, and can link to an Active Directory you already run on site. Pricing is per domain controller hour plus backup storage.","useCases":["Run Active Directory without managing servers","Host Windows apps that need Active Directory","Move Active Directory setups to cloud"],"pricing":"Per domain controller hour plus backup storage.","status":"ga","url":"https://cloud.google.com/managed-microsoft-ad","source":"https://cloud.google.com/products"},{"id":"gcp-identity-aware-proxy","provider":"gcp","name":"Identity-Aware Proxy","category":"zero-trust","also":["identity"],"tagline":"Let the right people reach internal apps without a VPN.","description":"Identity-Aware Proxy (IAP) checks who someone is, and the context of their request, before letting them reach your app. If you built an internal dashboard on Cloud Run or App Engine, you can limit it to your team without asking anyone to install a VPN, a private tunnel into your network. It also works for logging in to virtual machines over SSH (remote command-line access) when they have no public address. It covers apps on App Engine, Cloud Run, GKE and Compute Engine. It is free for most uses.","useCases":["Let staff reach internal apps","Protect internal apps without VPN","Open SSH to servers without VPN"],"pricing":"Free for most uses.","status":"ga","url":"https://cloud.google.com/security/products/iap","source":"https://cloud.google.com/products"},{"id":"gcp-access-context-manager","provider":"gcp","name":"Access Context Manager","category":"zero-trust","tagline":"Set rules for where and on what device people can reach your data.","description":"Access Context Manager lets you define access levels, which are rules about the situation someone is in when they ask for access. For example, you can say a request is trusted only if it comes from your office's IP range or from a device that meets your security checks. If you already use VPC Service Controls or Identity-Aware Proxy, those access levels plug straight into them. That way one set of rules can guard several things, instead of rewriting them each time. It is free.","useCases":["Define access by location or device","Set rules for corporate IP ranges","Allow access only from managed devices"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/access-context-manager/docs","source":"https://cloud.google.com/products"},{"id":"gcp-chrome-enterprise-premium","provider":"gcp","name":"Chrome Enterprise Premium","category":"zero-trust","tagline":"Secure access and data-leak protection built into the Chrome browser.","description":"Chrome Enterprise Premium adds security checks to the Chrome browser your team already uses. If your staff open sensitive customer data in web apps, it can scan for malware and apply data loss prevention (DLP), which stops private information from being pasted or shared where it shouldn't be. It also offers BeyondCorp-style secure access, Google's approach of trusting the person and their device rather than the office network. Access to apps works without installing an extra program (agentless). It is priced per user per month.","useCases":["Add data loss prevention in browsers","Scan downloads for malware","Secure app access without extra software"],"pricing":"Per user per month.","status":"ga","url":"https://docs.cloud.google.com/chrome-enterprise-premium/","source":"https://docs.cloud.google.com/chrome-enterprise-premium/"},{"id":"gcp-cloud-kms","provider":"gcp","name":"Cloud KMS","category":"secrets","tagline":"Create and manage the encryption keys that lock your data.","description":"Cloud KMS (Key Management Service) is where you create and control the encryption keys, the digital keys that scramble your data so only the right people can read it. If a customer or regulator wants you to hold the keys to your stored data, you can use it for customer-managed encryption keys (CMEK) on storage and databases. For stricter control you can add Cloud HSM, special tamper-resistant hardware, or External Key Manager, which keeps keys outside Google. It also signs and verifies data. Pricing is per key version per month plus per operation.","useCases":["Encrypt databases with your own keys","Create and manage encryption keys","Keep keys in tamper-resistant hardware"],"pricing":"Per key version per month plus per operation.","status":"ga","url":"https://cloud.google.com/security/products/security-key-management","source":"https://cloud.google.com/products"},{"id":"gcp-key-access-justifications","provider":"gcp","name":"Key Access Justifications","category":"secrets","tagline":"See why Google wants to use your encryption keys, then approve or deny.","description":"Key Access Justifications shows you the reason each time Google asks to use an encryption key that you store outside Google Cloud, and lets you approve or deny the request. It works with External Key Manager, which keeps your keys outside Google's systems. If you sell to banks, government or customers with strict rules about who can read their data, this gives you proof and control over every use of the key. Pick it when you need that level of control and a clear audit trail. It is part of the premium compliance offerings, so you will need to contact sales.","useCases":["Approve each use of your keys","Allow key use only with reasons","Audit when keys get used"],"pricing":"Part of premium compliance offerings; contact sales.","status":"ga","url":"https://cloud.google.com/assured-workloads/key-access-justifications/docs/overview","source":"https://cloud.google.com/products"},{"id":"gcp-secret-manager","provider":"gcp","name":"Secret Manager","category":"secrets","tagline":"Safely store API keys, passwords and certificates, with version history.","description":"Secret Manager is a locked vault for the sensitive bits your app needs, like API keys (the codes that let your app talk to other services), passwords and certificates. If your app uses a payment or AI service key, you can keep it here instead of pasting it into your code, and Cloud Run and GKE (Google's ways of running your app) can load it directly. It keeps older versions, logs who accessed each secret, and can copy secrets across regions. Pick it over a plain settings file when you want control over who can read what. There is a free monthly allowance.","useCases":["Store API keys and passwords safely","Rotate API keys on a schedule","Pass secrets into running apps"],"pricing":"Per active secret version plus access operations. Free monthly allowance.","status":"ga","url":"https://cloud.google.com/secret-manager","source":"https://cloud.google.com/products"},{"id":"gcp-certificate-manager","provider":"gcp","name":"Certificate Manager","category":"certificates","tagline":"Get and manage the HTTPS certificates that put the padlock on your sites.","description":"Certificate Manager handles TLS certificates, the files that turn on HTTPS and the padlock in the browser, for your Google Cloud load balancers (the front door that spreads visitors across your servers). If you run a product with many customer domains, you can attach lots of them to one load balancer. Google can issue certificates for you, or you can bring your own, and wildcard certificates that cover every subdomain are supported. Pick it when you want HTTPS on Google's global load balancers without handling each certificate by hand. You pay per certificate beyond a free allowance.","useCases":["Get HTTPS certificates for load balancers","Cover many domains with one setup","Issue wildcard certificates for subdomains"],"pricing":"Per certificate beyond a free allowance.","status":"ga","url":"https://cloud.google.com/certificate-manager","source":"https://cloud.google.com/products"},{"id":"gcp-certificate-authority-service","provider":"gcp","name":"Certificate Authority Service","category":"certificates","tagline":"Run your own private certificate authority without managing the hardware.","description":"Certificate Authority Service runs a private certificate authority, a system that hands out digital IDs to your own apps, servers and devices rather than to the public web. If you have internal services that need to prove who they are to each other, or a fleet of IoT devices (connected gadgets), this issues those IDs for you. The signing keys are protected in HSMs, special tamper-resistant hardware. Pick it over public HTTPS certificates when the certificates are only for internal use. You pay per CA each month plus per certificate issued.","useCases":["Run a private certificate authority","Issue certificates for internal services","Give devices certificate identities"],"pricing":"Per CA per month plus per certificate issued.","status":"ga","url":"https://cloud.google.com/certificate-authority-service","source":"https://cloud.google.com/products"},{"id":"gcp-security-command-center","provider":"gcp","name":"Security Command Center","category":"threat-intel","also":["management"],"tagline":"Spot risky settings, weak spots and active attacks across your cloud.","description":"Security Command Center is a dashboard that scans your cloud for risky settings, known weak spots and signs of an attack. If you have a growing set of projects and nobody has time to check each one by hand, it shows you what to fix first. It covers Google Cloud and other clouds, and can produce compliance reports. The Standard tier is free, so it is an easy first step. Premium and Enterprise add more, and Enterprise brings in SIEM and SOAR features (tools for collecting security logs and automating the response).","useCases":["Find security misconfigurations in cloud","Detect active threats across clouds","Generate compliance reports"],"pricing":"Standard tier free. Premium and Enterprise by usage or subscription.","status":"ga","url":"https://cloud.google.com/security/products/security-command-center","source":"https://cloud.google.com/products"},{"id":"gcp-sensitive-data-protection","provider":"gcp","name":"Sensitive Data Protection","category":"zero-trust","tagline":"Find personal data in your files and databases, then hide or mask it.","description":"Sensitive Data Protection, formerly called Cloud DLP, scans your data for personal information such as names and emails (often called PII) and can hide it. If your app stores user uploads or support messages and you want to train an AI model on them, you can mask or tokenize the sensitive parts first, meaning swap them for stand-in values. It works on BigQuery (Google's data warehouse), Cloud Storage and plain text. Pick it when you need to know where personal data is hiding in your systems. You pay per GB scanned or transformed.","useCases":["Find personal data in databases","Mask personal data before analysis","Scan storage for sensitive info"],"pricing":"Per GB scanned or transformed.","status":"ga","url":"https://cloud.google.com/security/products/sensitive-data-protection","source":"https://cloud.google.com/products"},{"id":"gcp-google-security-operations","provider":"gcp","name":"Google Security Operations","category":"threat-intel","tagline":"Collect security logs, search them fast and automate your response.","description":"Google Security Operations, formerly Chronicle, gathers security logs from across your systems into one place where you can search them quickly. It combines Chronicle SIEM (a searchable store of security logs) and Siemplify SOAR (a tool that automates response steps), and adds ready-made detections and Gemini-assisted investigation. If your security team needs to trace a suspicious login across many systems, this is where they would look. Pick it over a basic log viewer when you have huge volumes of security data to search. Pricing is through sales, based on data ingested or per employee.","useCases":["Search security logs quickly","Automate incident response steps","Hunt threats through log data"],"pricing":"Contact sales; ingest-based or per-employee pricing.","status":"ga","url":"https://cloud.google.com/security/products/security-operations","source":"https://cloud.google.com/products"},{"id":"gcp-google-threat-intelligence","provider":"gcp","name":"Google Threat Intelligence","category":"threat-intel","tagline":"Research hackers and malware using Mandiant, VirusTotal and Google data.","description":"Google Threat Intelligence puts data from Mandiant, VirusTotal and Google into one subscription so analysts can research who is attacking and how. If your security team spots a strange web address or file and wants to know whether known attackers are behind it, they can look up that indicator (an IOC, in security jargon) and read profiles of the adversary groups. It also analyzes malware and offers Gemini summaries. Pick it when you want answers from one source instead of checking several. It is sold in subscription tiers, so contact sales.","useCases":["Look up suspicious file hashes","Enrich alerts with threat data","Research known attacker groups"],"pricing":"Subscription tiers; contact sales.","status":"ga","url":"https://cloud.google.com/security/products/threat-intelligence","source":"https://cloud.google.com/products"},{"id":"gcp-mandiant-managed-defense","provider":"gcp","name":"Mandiant Managed Defense","category":"threat-intel","tagline":"A Mandiant expert team watches for attacks and responds, around the clock.","description":"Mandiant Managed Defense is a team of outside security experts who watch your systems around the clock, hunt for threats and step in when something looks wrong. If you do not have staff to review alerts at 3 a.m., this works as an outsourced security operations center, or SOC. The team draws on Mandiant's frontline experience with real attacks. Pick it over buying more security tools when what you are missing is people. Pricing comes from sales.","useCases":["Watch systems around the clock for attacks","Hunt for hidden threats in your setup","Respond to threats on your behalf"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/security/products/managed-defense","source":"https://cloud.google.com/products"},{"id":"gcp-mandiant-incident-response","provider":"gcp","name":"Mandiant Incident Response","category":"threat-intel","tagline":"Call in security experts to handle a breach and recover from it.","description":"Mandiant Incident Response gives you experienced responders who investigate a security breach, stop it from spreading and help you recover. If ransomware locks your files or you discover someone has been inside your systems, these are the people you call. You can also set up a retainer, a standing arrangement so help is lined up before anything goes wrong. Pick it when you need hands-on experts rather than another piece of software. Pricing comes from sales.","useCases":["Investigate and contain a security breach","Recover systems after ransomware attacks","Get expert help during a breach"],"pricing":"Contact sales; retainers.","status":"ga","url":"https://cloud.google.com/security/consulting/mandiant-incident-response-services","source":"https://cloud.google.com/products"},{"id":"gcp-mandiant-consulting","provider":"gcp","name":"Mandiant Consulting","category":"threat-intel","tagline":"Hire experts to test your defenses and plan your security program.","description":"Mandiant Consulting is a team of security experts you can hire to check how well your defenses hold up. They run assessments, tabletop exercises (practice runs of how your team would handle an attack) and help design cyber defense programs. If larger customers start asking hard security questions, a red team test, where experts try to break in on purpose, shows you where you are weak. Pick it when you want an outside view rather than grading your own work. Contact sales for pricing.","useCases":["Test defenses with simulated attacks","Practice breach response with tabletop exercises","Plan a long-term security program"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/security/consulting/mandiant-services","source":"https://cloud.google.com/products"},{"id":"gcp-fraud-defense","provider":"gcp","name":"Google Cloud Fraud Defense (reCAPTCHA)","category":"bots","tagline":"Block bots, fake accounts and SMS fraud on your forms and checkout.","description":"Google Cloud Fraud Defense is the next step beyond reCAPTCHA, the tool that tells real people from bots, and reCAPTCHA is now its bot defense layer. If your signup form gets flooded with fake accounts or bots keep hitting your checkout, it helps you spot and stop them. It also covers account takeover (someone logging in with stolen passwords), SMS toll fraud (bots running up your text message bills) and verification of AI agents. It launched in April 2026, and existing reCAPTCHA keys keep working. The first 10,000 assessments each month are free.","useCases":["Block bots on forms and checkout","Stop hackers from taking over user accounts","Verify AI agents interacting with your app"],"pricing":"Free tier for first 10,000 assessments per month, then per 1,000 assessments.","status":"ga","url":"https://cloud.google.com/security/products/fraud-defense","source":"https://cloud.google.com/products"},{"id":"gcp-web-risk","provider":"gcp","name":"Web Risk","category":"threat-intel","tagline":"Check links against Google's lists of unsafe websites.","description":"Web Risk checks web links against Google's lists of sites known for malware, phishing (fake pages that steal logins) and unwanted software. If your app lets people post links, send messages or share files, you can check each link before anyone clicks it. The same check works for scanning links inside emails. It uses Google's Safe Browsing data, so you don't have to build and maintain your own blocklist. You pay per 1,000 checks after a free allowance.","useCases":["Check links against known unsafe sites","Scan links in your app for malware","Flag phishing links in emails"],"pricing":"Per 1,000 API calls after a free allowance.","status":"ga","url":"https://cloud.google.com/web-risk","source":"https://cloud.google.com/products"},{"id":"gcp-binary-authorization","provider":"gcp","name":"Binary Authorization","category":"devtools","also":["threat-intel"],"tagline":"Only let approved, signed app packages go live.","description":"Binary Authorization is a gatekeeper that blocks app packages (called images) from going live unless they carry a signed approval, known as an attestation. If your team ships to Cloud Run or GKE, Google's service for running containers, you can set a rule so only packages that passed your checks get deployed. Unsigned ones are turned away. Pick it when you want a firm guarantee that an unreviewed build can't reach production. Pricing is based on policy-enforced attestations.","useCases":["Deploy only software that passes checks","Block unsigned software images from launching","Stop unverified builds from going live"],"pricing":"Per policy-enforced attestations; see pricing.","status":"ga","url":"https://cloud.google.com/binary-authorization","source":"https://cloud.google.com/products"},{"id":"gcp-artifact-analysis","provider":"gcp","name":"Artifact Analysis","category":"threat-intel","also":["devtools"],"tagline":"Scan your app packages for known security holes.","description":"Artifact Analysis scans the container images (packaged apps) and code libraries stored in Artifact Registry, Google's storage for packages, looking for known security flaws. It also records an SBOM, a parts list of everything inside your software, along with provenance, a record of where it came from. If you keep your packages in Artifact Registry, it flags risky ingredients before they reach customers. Pick it when you want scanning built into the place you already store your packages. Pricing is per image scanned.","useCases":["Scan software images for known vulnerabilities","Create a list of software components","Check language packages for security issues"],"pricing":"Per image scanned.","status":"ga","url":"https://cloud.google.com/artifact-analysis/docs","source":"https://cloud.google.com/products"},{"id":"gcp-software-supply-chain-security","provider":"gcp","name":"Software Supply Chain Security","category":"threat-intel","also":["devtools"],"tagline":"One view of how safely your app is built, stored and shipped.","description":"Software Supply Chain Security is a single view of the steps your code takes on the way to production: building, storing and deploying. It links Cloud Build records (Google's build service) of how something was made, Artifact Analysis scanning and Binary Authorization approval rules, then maps them to SLSA levels, an industry scale for how trustworthy a software build is. If a customer or auditor asks how you know your app wasn't tampered with, this gives you something to show. Choose it when you want those checks working together. You pay only for the components you use.","useCases":["Track how software was built and shipped","Prove software was built through trusted steps","Deliver builds only after checks pass"],"pricing":"Pay for components used.","status":"ga","url":"https://cloud.google.com/security/solutions/software-supply-chain-security","source":"https://cloud.google.com/products"},{"id":"gcp-assured-open-source-software","provider":"gcp","name":"Assured Open Source Software","category":"threat-intel","also":["devtools"],"tagline":"Open source Java and Python packages built and scanned by Google.","description":"Assured Open Source Software gives you popular free Java and Python packages that Google has built and scanned itself. They are served through Artifact Registry, Google's storage for packages, from Google's secure build pipeline. If your app pulls in many free libraries from Maven (for Java) or PyPI (for Python), you can use vetted versions instead of grabbing them straight from the public internet. Pick it when you want a more trusted source for the outside code your app relies on. Pricing is by contacting sales, or it is included with premium tiers.","useCases":["Use open source packages Google has scanned","Pull vetted Java and Python packages","Install safer dependencies for projects"],"pricing":"Contact sales / included with premium tiers.","status":"ga","url":"https://cloud.google.com/security/products/assured-open-source-software","source":"https://cloud.google.com/products"},{"id":"gcp-assured-workloads","provider":"gcp","name":"Assured Workloads","category":"management","tagline":"Compliance settings for regulated work like FedRAMP and CJIS.","description":"Assured Workloads applies the rules that governments and regulated industries require, such as FedRAMP (a US government cloud security standard) and CJIS (rules for US criminal justice data). It sets where your data lives, which Google staff can work on it and how it is encrypted, all on a folder in your Google Cloud account. If you are selling to government agencies, you can place that work in a folder that follows the right rules. Choose it when a contract demands compliance and you'd rather not build it yourself. It adds a premium percentage on usage, so contact sales.","useCases":["Keep government data in approved regions","Meet FedRAMP and CJIS rules","Set staff and encryption rules on folders"],"pricing":"Premium percentage on usage; contact sales.","status":"ga","url":"https://cloud.google.com/security/products/assured-workloads","source":"https://cloud.google.com/products"},{"id":"gcp-audit-manager","provider":"gcp","name":"Audit Manager","category":"management","tagline":"Collect evidence for compliance audits and generate reports.","description":"Audit Manager helps you prepare for a compliance audit, the formal check that you follow the security rules a customer or regulator expects. It matches those rules to your Google Cloud resources, collects the evidence and generates audit-ready reports. If a big customer asks you to prove how you protect their data, you can hand over a report instead of assembling it by hand. Pick it when audits are coming and you want the evidence gathered in one place. Pricing is usage-based.","useCases":["Collect proof for compliance audits","Map security controls to your resources","Generate reports ready for auditors"],"pricing":"Usage-based; see pricing.","status":"ga","url":"https://docs.cloud.google.com/audit-manager/docs","source":"https://docs.cloud.google.com/audit-manager/docs"},{"id":"gcp-access-approval","provider":"gcp","name":"Access Approval","category":"management","tagline":"Say yes or no when Google staff want to access your data.","description":"Access Approval makes Google support staff ask for your explicit consent before they can access your content. If you store sensitive customer data, such as in a regulated field, you can be sure no one at Google looks at it without your say. Pick it when your customers or regulators expect you to control who touches their data, rather than just trusting a promise. It is included with premium support or security tiers.","useCases":["Require your approval before Google staff see data","Keep support access to customer data under your control","Meet consent rules for regulated customer data"],"pricing":"Included with premium support/security tiers.","status":"ga","url":"https://cloud.google.com/access-approval","source":"https://cloud.google.com/products"},{"id":"gcp-access-transparency","provider":"gcp","name":"Access Transparency","category":"management","tagline":"See a log of when Google staff access your content.","description":"Access Transparency gives you near real-time logs of when Google administrators access your content. If a customer asks who has looked at their data, or an auditor wants a trail, you have a record to show. It pairs well with Access Approval: approval lets you say yes or no beforehand, while transparency shows what actually happened. Pick it when you want visibility without approving each request. It is included with qualifying support packages.","useCases":["Log admin actions by Google staff","Review provider access to your content","Keep an audit trail of provider access"],"pricing":"Included with qualifying support packages.","status":"ga","url":"https://cloud.google.com/access-transparency","source":"https://cloud.google.com/products"},{"id":"gcp-advisory-notifications","provider":"gcp","name":"Advisory Notifications","category":"management","tagline":"Get notices about security and privacy issues affecting you.","description":"Advisory Notifications sends you notices from Google about security and privacy events that affect your account. It focuses on sensitive advisories and keeps unrelated alerts down, so what lands in your inbox is worth reading. If you are a small team without a dedicated security person, it helps make sure the right people hear about problems that matter, since you can route notices to your security contacts. Pick it over general alert emails when you want less noise. It is free.","useCases":["Receive security and privacy notices that matter","Route security alerts to the right contact","Get fewer unrelated alerts"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/advisory-notifications","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-asset-inventory","provider":"gcp","name":"Cloud Asset Inventory","category":"management","tagline":"A searchable list of everything in your cloud, with change history.","description":"Cloud Asset Inventory is a searchable list of everything in your Google Cloud account, including your resources, who has access to them (called IAM policies) and what changed over time. If you ever wonder what your team set up and who changed it, you can search instead of clicking through screens. You can also export the data to BigQuery, Google's tool for analyzing large tables of data, to build compliance reports. It is free, which makes it an easy first step for keeping track of a growing project.","useCases":["Search all your cloud resources at once","Track changes to resources over time","Export the inventory for compliance reports"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/asset-inventory/docs","source":"https://cloud.google.com/products"},{"id":"gcp-policy-intelligence","provider":"gcp","name":"Policy Intelligence","category":"management","also":["identity"],"tagline":"Find out who has access to what, and trim what's unused.","description":"Policy Intelligence is a set of tools that help you understand and tighten who can do what in your cloud account. Policy Analyzer, Troubleshooter and Recommender show who has access, help explain why someone can or can't do something, and point out permissions nobody uses. If your team has grown and everyone has broad access, you can cut back to only what people need, a practice called least privilege. Choose it when you want to tidy up access using evidence instead of guesses. It has a free tier, and premium tools are priced by usage.","useCases":["Find who has access to what","Remove permissions nobody uses","Troubleshoot why access is denied"],"pricing":"Free tier; premium tools by usage.","status":"ga","url":"https://cloud.google.com/policy-intelligence","source":"https://cloud.google.com/products"},{"id":"gcp-resource-manager","provider":"gcp","name":"Resource Manager","category":"management","tagline":"Organize your Google Cloud projects into folders and set rules at the top.","description":"Resource Manager is how you organize everything you run on Google Cloud. Projects (separate containers for each app and its bill) get grouped into folders under one organization. If you have a live product, a test version and an internal tool, you can give each its own project and let your team into only the ones they need. Permissions and policies set at the top flow down to everything underneath, so you do not repeat yourself. It is free, and it is worth setting up early, before your projects multiply.","useCases":["Organize projects into folders and groups","Apply access rules from the top down","Manage the life of each project"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/resource-manager","source":"https://cloud.google.com/products"},{"id":"gcp-organization-policy-service","provider":"gcp","name":"Organization Policy Service","category":"management","tagline":"Set rules for your whole cloud account, like where your data can live.","description":"Organization Policy Service lets you set guardrails that apply across your folders and projects in Google Cloud. For example, you can allow servers to be created only in certain regions (geographic locations), or ban public IP addresses, which are internet-facing addresses that can expose a machine by accident. If several people or AI tools create resources for you, it stops mistakes before they happen instead of catching them later. It is free, and it follows the folder structure you set up in Resource Manager.","useCases":["Limit which regions resources can use","Block public IP addresses across projects","Set guardrails for all folders"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/resource-manager/docs/organization-policy/overview","source":"https://cloud.google.com/products"},{"id":"gcp-sovereign-controls-by-partners","provider":"gcp","name":"Sovereign Controls by Partners","category":"management","also":["secrets"],"tagline":"Keep data in your country, with a local partner holding your encryption keys.","description":"Sovereign Controls by Partners is a Google Cloud option where a local partner company operates the security controls and holds your encryption keys, the secret codes that lock your data. The goal is to keep your data in-country. If you sell to governments, banks or healthcare providers in Europe and beyond, and they require local control, this helps you meet that. Pick it over standard settings when the question is not only where your data sits but who controls access to it. Pricing is through contacting sales.","useCases":["Keep data inside your country","Have local partners hold encryption keys","Run controls for European data"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/sovereign-controls-by-partners","source":"https://cloud.google.com/products"},{"id":"gcp-unified-maintenance","provider":"gcp","name":"Unified Maintenance","category":"management","tagline":"One place to see and plan when Google Cloud updates and patches happen.","description":"Unified Maintenance gives you one view of upcoming maintenance across Google Cloud services, such as updates and patches (fixes for bugs and security holes). If your app relies on a database, servers and a few other services, each might have its own schedule, and this brings them into a single picture so you can plan around them. Pick it when you want to avoid surprise disruption during a launch or a busy sales period. It is free.","useCases":["See all planned maintenance in one view","Schedule updates across Google Cloud services","Plan patches around your team's schedule"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/unified-maintenance/docs","source":"https://docs.cloud.google.com/unified-maintenance/docs"},{"id":"gcp-service-health","provider":"gcp","name":"Service Health","category":"observability","also":["management"],"tagline":"Find out when a Google Cloud outage affects your app, with alerts.","description":"Service Health tells you when a Google Cloud problem is affecting the things you actually use. If your app suddenly slows down or goes offline, you can check here to see whether it is a Google incident or a bug in your own code. It shows only incidents relevant to your resources and can send you alerts, so you do not have to keep watching a general status page. It is free.","useCases":["Track Google Cloud outages affecting you","Get alerts about incidents on your resources","See whether a provider outage is the cause"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/service-health/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-hub","provider":"gcp","name":"Cloud Hub","category":"management","tagline":"One dashboard for your apps, their health and what they cost.","description":"Cloud Hub is a single console page that shows your applications, how healthy they are and what they cost across your Google Cloud setup. If you are a founder who wants a quick answer to whether things are running and where the money is going, without opening ten separate tools, this is the place to start. Treat it as your overview, then jump into specific tools when you need detail. It is free.","useCases":["See apps, health and costs in one place","Monitor your whole cloud setup from one console","Check deployments and costs across all resources"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/hub/docs","source":"https://docs.cloud.google.com/cloud-hub/docs"},{"id":"gcp-cloud-api-registry","provider":"gcp","name":"Cloud API Registry","category":"management","also":["agents"],"tagline":"A directory of Google Cloud APIs and MCP servers, for people and AI agents.","description":"Cloud API Registry is a catalog where you can find Google Cloud APIs (the doors your software uses to talk to other services) and MCP endpoints, which are connection points that let AI agents use tools. If you are building an AI agent that needs to discover what it can call, this gives you one place to look instead of hunting through documentation. It is currently in preview, meaning it is still being tested and may change. Check the docs for pricing.","useCases":["Find cloud APIs for AI agents to use","Look up available APIs in one catalog","Discover tool servers that agents can call"],"pricing":"Preview; see docs.","status":"preview","url":"https://docs.cloud.google.com/api-registry/docs","source":"https://docs.cloud.google.com/api-registry/docs"},{"id":"gcp-google-cloud-observability","provider":"gcp","name":"Google Cloud Observability","category":"observability","tagline":"One home for logs, metrics, traces and error tracking for your app.","description":"Google Cloud Observability, formerly known as Operations Suite and Stackdriver, is the bundle of tools that tell you what your app is doing and why it breaks. It brings together logging, monitoring, tracing, profiling and error tools, and it accepts OpenTelemetry, a common open standard for sending app data. If your app goes down at night, you can investigate in one place, with Gemini offering troubleshooting help. You pay per product, so check each individual entry for prices.","useCases":["Keep logs, metrics and traces in one place","Collect app telemetry using OpenTelemetry","Troubleshoot problems with Gemini assistance"],"pricing":"Pay per product; see individual entries.","status":"ga","url":"https://cloud.google.com/products/observability","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-logging","provider":"gcp","name":"Cloud Logging","category":"observability","tagline":"Collect, search and analyze your app's logs in one place.","description":"Cloud Logging collects the running diary your apps and Google Cloud services write, called logs, and lets you search it all in one place. If a customer says checkout failed, you can look up what happened at that moment instead of guessing. You can also analyze logs with Log Analytics in BigQuery, Google's data analysis tool, and set alerts based on what appears in them. You pay per GiB (a unit of data) ingested, and the first 50 GiB per project each month are free.","useCases":["Store and search app logs centrally","Alert when specific log entries appear","Keep audit logs for later review"],"pricing":"Per GiB ingested. First 50 GiB per project free monthly.","status":"ga","url":"https://cloud.google.com/logging","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-monitoring","provider":"gcp","name":"Cloud Monitoring","category":"observability","tagline":"Dashboards and alerts that tell you when your app is down or slow.","description":"Cloud Monitoring tracks numbers about your app, called metrics, such as how busy your servers are, and shows them on dashboards. If your site goes down at 3 a.m., uptime checks (automatic tests that visit your site) and alert policies can notify you before customers complain. You can also set SLOs, which are targets for how reliable you want to be, and run queries in the style of Prometheus, a popular open source monitoring tool. You pay per MiB of metrics ingested, with a free allotment.","useCases":["Watch app performance on live dashboards","Get alerted when something breaks","Check that websites stay online"],"pricing":"Per MiB of metrics ingested; free allotment.","status":"ga","url":"https://cloud.google.com/monitoring","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-trace","provider":"gcp","name":"Cloud Trace","category":"observability","tagline":"Follow a request through your app to find out where it slows down.","description":"Cloud Trace shows the path a single request takes through your app and how long each step takes. If a page feels slow and your app is made of several connected services, it helps you spot which one is the holdup. It collects this data using OpenTelemetry, a common open standard, and shows how response times are spread out. It is most useful once you have more than one service talking to each other. You pay per million spans (individual timed steps) after a free allowance.","useCases":["Find which requests load slowly","Follow one request across several services","Compare response times across services"],"pricing":"Per million spans after a free allowance.","status":"ga","url":"https://cloud.google.com/trace/docs","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-profiler","provider":"gcp","name":"Cloud Profiler","category":"observability","tagline":"Find which parts of your code use the most computing power, for free.","description":"Cloud Profiler watches your live app and shows which parts of your code use the most processing power and memory. If your cloud bill is climbing or your app feels sluggish, it points to the slow, expensive pieces so you know where to fix first. Results appear as flame graphs, charts where wider bars mean more time spent. It is built to run all the time without big slowdowns for your users. It is free.","useCases":["Find code that wastes computing power","Cut server bills by fixing slow code","Study live app speed without slowing it"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/profiler/docs","source":"https://cloud.google.com/products"},{"id":"gcp-error-reporting","provider":"gcp","name":"Error Reporting","category":"observability","tagline":"Spot app crashes, grouped together, and get told when a new one appears.","description":"Error Reporting watches your app's logs for crashes and groups the same error together, so you see one problem instead of thousands of repeats. If you just shipped an update and a new bug starts hitting users, it notifies you about the new error, so you don't have to dig through logs to find out. It's a good first stop for seeing what broke and how often. It costs nothing to use.","useCases":["Group repeated crashes into one list","Get notified when a new error appears","Track errors found in app logs"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/error-reporting/docs/grouping-errors","source":"https://cloud.google.com/products"},{"id":"gcp-managed-service-for-prometheus","provider":"gcp","name":"Managed Service for Prometheus","category":"observability","tagline":"Track how your app is running with Prometheus, without hosting it yourself.","description":"Managed Service for Prometheus stores the running numbers (metrics) that Prometheus, a popular open-source monitoring tool, collects about your app. If you run your app on Kubernetes, a system for running many app pieces at once, you can keep its metrics here and look at them in Grafana dashboards. You ask questions with PromQL, Prometheus's own query language. Pick it over running Prometheus yourself when you don't want to manage storage servers. You pay per sample ingested.","useCases":["Collect metrics from container clusters","Feed metrics into Grafana dashboards","Keep metrics without running servers"],"pricing":"Per sample ingested.","status":"ga","url":"https://cloud.google.com/managed-prometheus","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-billing","provider":"gcp","name":"Cloud Billing","category":"management","tagline":"See what you spend on Google Cloud, set budgets and get alerts.","description":"Cloud Billing is where your invoices, budgets and spending alerts live. If you're worried an experiment or a busy weekend could produce a surprise bill, set a budget and get an alert as costs climb. You can also export your billing data to BigQuery, Google's tool for analyzing large amounts of data, to build your own cost reports. It's the place to start for understanding your spend, and it's free.","useCases":["Set spending budgets with alerts","Review monthly cloud cost reports","Export billing data for deeper analysis"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/billing","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-quotas","provider":"gcp","name":"Cloud Quotas","category":"management","tagline":"See your usage limits on Google Cloud and ask for higher ones.","description":"Cloud Quotas is one place to check the limits Google puts on how much of each service you can use, and to ask for changes. If you're about to train an AI model and need more GPUs, the chips that do that heavy work, you can request a higher GPU limit here before you hit a wall. It saves you from hunting through each service for its own limit settings. It's free to use.","useCases":["Check usage limits for each service","Request higher GPU limits","Manage service limits from one place"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/docs/quotas","source":"https://cloud.google.com/products"},{"id":"gcp-carbon-footprint","provider":"gcp","name":"Carbon Footprint","category":"management","tagline":"See the carbon emissions created by your Google Cloud usage.","description":"Carbon Footprint shows the emissions from your cloud usage, both gross and net, broken down by project and by service. If a customer or investor asks for sustainability numbers, you can pull them from here instead of estimating. It also helps you see which choices matter, such as which region (the physical location of Google's data centers) your app runs in, if you want to cut emissions. It's free.","useCases":["Report emissions for each project","Choose regions with lower emissions","See emissions broken down by service"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/carbon-footprint","source":"https://cloud.google.com/products"},{"id":"gcp-recommender","provider":"gcp","name":"Recommender (Active Assist)","category":"management","tagline":"Get tips to cut cloud costs, tighten security and clear out waste.","description":"Recommender, part of Active Assist, looks at how you use Google Cloud and suggests improvements. It flags idle resources you're paying for but not using, servers that are bigger than they need to be, and security suggestions. If a teammate left months ago and still has access, it can point that out so you can clean up permissions. It's a quick way to find savings without auditing everything by hand. There's a free tier, and premium insights cost extra.","useCases":["Find idle resources to shut down","Get right-sizing tips for servers","Clean up unused access permissions"],"pricing":"Free tier; premium insights extra.","status":"ga","url":"https://cloud.google.com/recommender/docs/whatis-activeassist","source":"https://cloud.google.com/products"},{"id":"gcp-fault-injection-testing","provider":"gcp","name":"Fault Injection Testing","category":"observability","also":["devtools"],"tagline":"Break things on purpose to check your app recovers.","description":"Fault Injection Testing lets you simulate failures in your services so you can see whether your app recovers the way you expect. If your product has to stay up during a busy launch, you can rehearse what happens when a piece goes down, in a practice session often called a game day. That beats finding out during a real outage. It is still in preview, so check the docs for current details and pricing.","useCases":["Simulate failures to test recovery","Run game day resilience drills","Check services recover from outages"],"pricing":"Preview; see docs.","status":"preview","url":"https://docs.cloud.google.com/fault-injection-testing","source":"https://docs.cloud.google.com/fault-injection-testing/docs"},{"id":"gcp-migration-center","provider":"gcp","name":"Migration Center","category":"migration","tagline":"Plan and track your move to Google Cloud in one place.","description":"Migration Center helps you assess, plan and run a move to Google Cloud. If you have a collection of servers and apps running somewhere else, it discovers what you have, estimates what it would cost to move, and tracks progress along the way. Use it before you commit, so the budget and the order of work aren't guesses. It's free to use.","useCases":["Plan a move to Google Cloud","Estimate what moving will cost","Track progress of each migration"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/migration-center/docs","source":"https://docs.cloud.google.com/migration-center/docs"},{"id":"gcp-migrate-to-virtual-machines","provider":"gcp","name":"Migrate to Virtual Machines","category":"migration","also":["vms"],"tagline":"Move your existing virtual servers to Google Cloud with little downtime.","description":"Migrate to Virtual Machines moves VMs, which are virtual computers, into Compute Engine, Google's service for running virtual servers. If your app runs on VMware, AWS or Azure and you want to switch to Google Cloud, it copies the machines over with minimal downtime. This approach is often called lift and shift, because you move things as they are instead of rebuilding them. It's free for migrations to Compute Engine.","useCases":["Move virtual servers with little downtime","Shift VMware machines to Google Cloud","Copy AWS or Azure servers over"],"pricing":"Free for migrations to Compute Engine.","status":"ga","url":"https://cloud.google.com/products/cloud-migration/virtual-machines","source":"https://cloud.google.com/products"},{"id":"gcp-migrate-to-containers","provider":"gcp","name":"Migrate to Containers","category":"migration","also":["containers"],"tagline":"Turn your old virtual servers into containers that run on GKE.","description":"Migrate to Containers analyzes an app running on a virtual machine and generates the files needed to package it as a container, a self-contained bundle that runs the same way anywhere. If you have an older app on a VM and want it on GKE, Google's managed Kubernetes service, this gives you a head start over rebuilding it. Choose it when you want to modernize an existing app rather than just move it. It's free.","useCases":["Turn old server apps into containers","Modernize older apps for container hosting","Analyze server apps and build containers"],"pricing":"Free.","status":"ga","url":"https://cloud.google.com/products/cloud-migration/containers","source":"https://cloud.google.com/products"},{"id":"gcp-database-migration-service","provider":"gcp","name":"Database Migration Service","category":"migration","also":["sql"],"tagline":"Move your database to Google Cloud while it keeps running.","description":"Database Migration Service moves your database into Cloud SQL, AlloyDB or Spanner, all managed databases on Google Cloud. It keeps copying changes from your source, which can be MySQL, PostgreSQL, SQL Server or Oracle, so your app stays online until you switch over. If you're moving from Oracle to PostgreSQL, it supports that too. It's free when the database type stays the same, and the price varies when it changes.","useCases":["Move databases with minimal downtime","Move Oracle databases to PostgreSQL","Keep databases synced while migrating"],"pricing":"Free for homogeneous; heterogeneous varies.","status":"ga","url":"https://cloud.google.com/database-migration","source":"https://cloud.google.com/products"},{"id":"gcp-dual-run","provider":"gcp","name":"Dual Run","category":"migration","tagline":"Run old mainframe workloads in the cloud side by side to check results.","description":"Dual Run lets you run a mainframe workload, the big legacy computers that often power banks and insurers, in parallel with a cloud version. If you're moving a core banking system, you can compare the results from both before you switch over, so you find mistakes while the old system is still in charge. Pick it when a move is too risky to do on faith. Pricing is by contacting Google's sales team.","useCases":["Run mainframe apps alongside new systems","Compare old and new results before switching","Validate core banking systems before switching"],"pricing":"Contact sales.","status":"ga","url":"https://docs.cloud.google.com/mainframe-dual-run/docs","source":"https://docs.cloud.google.com/dual-run/docs"},{"id":"gcp-mainframe-connector","provider":"gcp","name":"Mainframe Connector","category":"migration","also":["data-pipelines"],"tagline":"Copy old mainframe data into Google's analytics and file storage.","description":"Mainframe Connector moves data out of mainframes, the large legacy computers many big companies still run, and converts it into formats modern tools can read. If your company has years of records on a mainframe and you want to analyze them, you can send them to BigQuery, Google's data analysis tool, or to Cloud Storage, its file storage. It also works for scheduled batch jobs that write results straight into BigQuery. Pick it when you want to keep the mainframe running but get its data into modern tools. Pricing is through Google's sales team.","useCases":["Move mainframe data into BigQuery","Run batch jobs that write to BigQuery","Offload mainframe data to cloud storage"],"pricing":"Contact sales.","status":"ga","url":"https://docs.cloud.google.com/mainframe-connector/docs","source":"https://docs.cloud.google.com/mainframe-connector/docs"},{"id":"gcp-mainframe-assessment-tool","provider":"gcp","name":"Mainframe Assessment Tool","category":"migration","tagline":"Scan your mainframe code to see how hard moving it would be.","description":"Mainframe Assessment Tool reads through the code on an old mainframe, the large legacy computer some big companies still depend on, and shows your options for modernizing it. If you are trying to decide whether to move a mainframe system to the cloud, it helps you estimate how much work the move would take before you commit a budget. It is a planning step rather than a migration tool, so you use it first to get a clear picture. It is free, which makes it an easy place to start.","useCases":["Assess how ready a mainframe is","Estimate effort to modernize old code","Review mainframe code for upgrade options"],"pricing":"Free.","status":"ga","url":"https://docs.cloud.google.com/mainframe-assessment-tool/docs","source":"https://docs.cloud.google.com/mainframe-assessment-tool/docs"},{"id":"gcp-live-stream-api","provider":"gcp","name":"Live Stream API","category":"video","tagline":"Turn a live camera feed into video people can watch in a browser.","description":"Live Stream API takes a live video feed from your camera or encoder and converts it into formats that play on phones, browsers and smart TVs. If you are running a live event or a sports broadcast, you send in your stream using RTMP or SRT, which are common ways to send live video. It then outputs HLS and DASH, the standard formats for streaming to viewers. Choose it when you want Google to handle the conversion instead of running video servers yourself. You pay per minute of video it outputs.","useCases":["Broadcast live events to viewers","Stream live sports to viewers","Convert live video into web formats"],"pricing":"Per minute of output.","status":"ga","url":"https://cloud.google.com/livestream/docs","source":"https://cloud.google.com/products"},{"id":"gcp-transcoder-api","provider":"gcp","name":"Transcoder API","category":"video","tagline":"Convert video files so they play smoothly on any device or connection.","description":"Transcoder API takes video files you have already recorded and converts them into versions that play well on the web. If your app lets people upload videos, or you run a library of lessons or shows, you can use it to create several quality levels, so viewers on slow connections still get smooth playback. This is called an adaptive bitrate ladder, meaning the player switches quality as the connection changes. Pick it for recorded video, and use Live Stream API for live events. You pay per minute of video it outputs.","useCases":["Convert uploaded videos for web playback","Create multiple quality versions of videos","Batch process large video libraries"],"pricing":"Per minute of output.","status":"ga","url":"https://cloud.google.com/transcoder/docs","source":"https://cloud.google.com/products"},{"id":"gcp-video-stitcher-api","provider":"gcp","name":"Video Stitcher API","category":"video","tagline":"Insert ads into your videos and live streams before they reach viewers.","description":"Video Stitcher API weaves ads into your video so they play as part of one continuous stream. If you run a streaming site or app and want to earn money from ads, you can use it to place ad breaks in live streams or recorded videos. It can also show different ads to different viewers. Because the ads are added on the server rather than in the viewer's player, they are harder for ad blockers to skip. It works for both live and on-demand video. You pay per minute of stitched video.","useCases":["Insert ads into live streams","Personalize ad breaks for each viewer","Add ads to on-demand videos"],"pricing":"Per minute of stitched video.","status":"ga","url":"https://cloud.google.com/video-stitcher/docs","source":"https://cloud.google.com/products"},{"id":"gcp-google-distributed-cloud","provider":"gcp","name":"Google Distributed Cloud","category":"containers","also":["vms"],"tagline":"Run Google Cloud services in your own data center or at the edge.","description":"Google Distributed Cloud brings Google's cloud technology to hardware that sits in your own data center or at remote edge locations, instead of in Google's buildings. If you have rules that require your data to stay in a certain place, or you want to run AI close to where data is created, it lets you do that. It can stay connected to Google Cloud or run fully air-gapped, meaning cut off from the internet. Choose it when the regular cloud is not allowed or not practical. Pricing is through Google's sales team.","useCases":["Run cloud services in your own data center","Run AI at edge locations","Run Google cloud services on local hardware"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/distributed-cloud","source":"https://cloud.google.com/products"},{"id":"gcp-google-distributed-cloud-connected","provider":"gcp","name":"Google Distributed Cloud Connected","category":"containers","also":["edge-compute"],"tagline":"Google-managed servers placed at your sites and connected to Google Cloud.","description":"Google Distributed Cloud Connected is hardware, such as racks of servers, that Google manages for you at your own location or at a telecom site. You control it from Google Cloud, so there is no need to run the machines yourself. If you run a telecom network or a chain of stores that wants AI to work on-site, for example analyzing camera video in each store, it keeps the processing close to where things happen. Pick it over a regular cloud region when speed and location matter. Pricing is through Google's sales team.","useCases":["Run managed servers at retail stores","Power edge AI at telecom sites","Manage hardware at remote customer sites"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/distributed-cloud-connected","source":"https://cloud.google.com/products"},{"id":"gcp-google-distributed-cloud-air-gapped","provider":"gcp","name":"Google Distributed Cloud Air-gapped","category":"containers","tagline":"A Google Cloud that runs with no internet connection at all.","description":"Google Distributed Cloud Air-gapped is a version of Google's cloud that is fully isolated from the internet, built for governments and defense organizations. If your team handles classified work or intelligence data that must never touch a public network, it gives you modern cloud tools inside a sealed environment. Air-gapped means physically separated from outside networks. Choose it when security rules rule out the regular cloud completely. For most startups and app builders it will be more than you need. Pricing is through Google's sales team.","useCases":["Run cloud fully offline for sensitive data","Host classified government workloads","Support defense and intelligence systems"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/distributed-cloud-air-gapped","source":"https://cloud.google.com/products"},{"id":"gcp-cloud-healthcare-api","provider":"gcp","name":"Cloud Healthcare API","category":"specialized","tagline":"Store and share medical records and scans in standard health formats.","description":"Cloud Healthcare API is a managed place to keep health data in the formats hospitals and clinics already use: FHIR and HL7v2 for patient records and messages, and DICOM for medical images like X-rays. If you are building a health app that needs to read records or store scans, it saves you from building that storage yourself. The data can then flow into BigQuery, Google's data analysis tool, and into AI tools. Pick it when you need to work with standard medical data formats. You pay per request and for storage.","useCases":["Store and exchange health records","Manage medical imaging files","Feed health data into analytics and AI"],"pricing":"Per request and storage.","status":"ga","url":"https://cloud.google.com/healthcare-api","source":"https://cloud.google.com/products"},{"id":"gcp-earth-engine","provider":"gcp","name":"Earth Engine","category":"specialized","tagline":"Analyze satellite images of the whole planet without owning the data.","description":"Earth Engine gives you a huge collection of satellite imagery and the computing power to analyze it. If you want to map how land is used, track forests over time or study climate trends, you can run your analysis there instead of downloading enormous image files. Geospatial means anything tied to places on the map. It is a good pick when your question is about the Earth itself. It is free for noncommercial use, such as research and education, and paid for commercial projects.","useCases":["Map land use from satellite images","Analyze climate trends from satellite data","Analyze large geographic datasets"],"pricing":"Free noncommercial; paid commercial.","status":"ga","url":"https://cloud.google.com/earth-engine","source":"https://cloud.google.com/products"},{"id":"gcp-device-connect-for-fitbit","provider":"gcp","name":"Device Connect for Fitbit","category":"specialized","tagline":"Bring Fitbit wearable health data into Google Cloud for analysis.","description":"Device Connect for Fitbit sets up the data pipelines that move Fitbit information, such as activity and sleep, into Google Cloud so you can analyze it. If you run a health study or a wellness program for employees or customers, it helps you collect that data once people agree to share it. It is aimed at health programs rather than general apps. There is no separate fee for it. You pay for the Google Cloud services that store and process the data.","useCases":["Study health data from Fitbit devices","Track wellness program results over time","Move Fitbit data into analytics tools"],"pricing":"Pay for underlying services.","status":"ga","url":"https://cloud.google.com/device-connect","source":"https://cloud.google.com/products"},{"id":"gcp-telecom-network-automation","provider":"gcp","name":"Telecom Network Automation","category":"specialized","tagline":"Automatically set up and manage software that runs telecom networks.","description":"Telecom Network Automation is a managed service that automates how a phone network's software pieces, called network functions, get deployed and updated. If you operate a 5G network, it handles the repetitive work of rolling out and maintaining those pieces across many sites. It is based on Nephio, an open source project for automating telecom networks. Pick it when you want that automation managed for you instead of building it yourself. It is built for telecom companies, not typical app builders. Pricing is through Google's sales team.","useCases":["Run 5G network operations","Deploy phone network functions automatically","Manage network functions from start to finish"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/telecom-network-automation","source":"https://cloud.google.com/products"},{"id":"gcp-telecom-data-fabric","provider":"gcp","name":"Telecom Data Fabric","category":"specialized","also":["data-pipelines"],"tagline":"Ready-made data pipelines that bring a phone company's data together.","description":"Telecom Data Fabric is a set of ready-made blueprints for moving and combining the data a phone or internet provider collects. If you build tools for a carrier and need to study how the network is performing, you can start from these pipelines (automated routes that move data from one place to another) instead of designing them yourself. It also helps when you need to join network operations data (OSS) with billing and customer data (BSS) in one place. Pick it over building from scratch when your data comes from telecom systems. Pricing is through Google's sales team.","useCases":["Analyze network traffic and performance","Combine network and billing data","Pull telecom data into one place"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/telecom-data-fabric","source":"https://cloud.google.com/products"},{"id":"gcp-telecom-subscriber-insights","provider":"gcp","name":"Telecom Subscriber Insights","category":"specialized","tagline":"Learn about a carrier's customers and tailor offers to them.","description":"Telecom Subscriber Insights gives phone and internet providers a set of APIs (ways for software to talk to software) for understanding their subscribers. If you build a product for a carrier and want to spot customers likely to cancel, you can use it to see who is at risk and decide which offer to send them. It is also useful for studying subscriber behavior in general. Pick it over a generic analytics tool when your customers are communications providers and you want telecom-specific tooling. Pricing is through Google's sales team.","useCases":["Analyze customer usage for carriers","Spot customers likely to leave","Target offers to the right subscribers"],"pricing":"Contact sales.","status":"ga","url":"https://cloud.google.com/telecom-subscriber-insights","source":"https://cloud.google.com/products"},{"id":"gcp-appsheet","provider":"gcp","name":"AppSheet","category":"end-user","tagline":"Turn spreadsheets and databases into mobile and web apps without code.","description":"AppSheet is a no-code app builder that turns your spreadsheets and databases into working mobile and web apps. If your team tracks inventory, inspections or customer visits in Google Sheets, you can turn that sheet into an app people fill in from their phones in the field. It also lets you add automation, such as steps that run when a new row appears. Pick it over writing an app yourself when you want an internal tool fast and your data already lives in Google Workspace or a database. It is free to prototype, then billed per user per month.","useCases":["Build internal tools from spreadsheets","Create mobile apps for field teams","Automate workflows without writing code"],"pricing":"Free to prototype; per user per month.","status":"ga","url":"https://cloud.google.com/appsheet/automation","source":"https://cloud.google.com/products"},{"id":"gcp-firebase-authentication","provider":"gcp","name":"Firebase Authentication","category":"identity","tagline":"Add sign-in to your app with email, Google, Apple, phone or guest access.","description":"Firebase Authentication handles the sign-in part of your app so you do not have to build it. If your app needs people to create accounts, you can let them log in with a password, their Google or Apple account, or a phone number, and you can even allow guest visitors. It comes with ready-made code kits (SDKs) and a hosted sign-in screen. Pick it over building your own login when you want something quick and safe to set up. Email and social sign-in are free, while phone text messages and advanced features such as company single sign-on (through Identity Platform) are billed.","useCases":["Add email, Google and Apple sign-in","Let users sign in by phone","Allow guest users without accounts"],"pricing":"Free for email/social. Phone SMS and Identity Platform features billed.","status":"ga","url":"https://firebase.google.com/products/auth","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-firestore","provider":"gcp","name":"Cloud Firestore (Firebase)","category":"realtime-db","also":["nosql"],"tagline":"A database for your app that updates live and works offline.","description":"Cloud Firestore is a database that stores your app's information as documents (think small records, like one per user) and scales on its own as you grow. If you are building a chat or team app, it pushes new messages to everyone's screen as they arrive, and it keeps working when someone loses signal, syncing later. It is the same service as Google Cloud Firestore, with extra Firebase tools such as security rules that control who can read what. Pick it for apps that need live updates without running your own servers. You pay per read, write, delete and storage, with a daily free allowance on the Spark plan.","useCases":["Build chat and collaboration apps","Keep apps working offline on phones","Store user profile information"],"pricing":"Per read, write, delete and storage. Daily free quota on Spark plan.","status":"ga","url":"https://firebase.google.com/products/firestore","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-realtime-database","provider":"gcp","name":"Firebase Realtime Database","category":"realtime-db","also":["nosql"],"tagline":"The original Firebase database that keeps every user's screen in sync.","description":"Firebase Realtime Database stores your data as one big JSON tree (a nested list of labeled values) and keeps it in sync across everyone using your app, almost instantly. If you want a live scoreboard or a green dot showing who is online, it handles that with very little setup, and it keeps working offline. Rules based on the signed-in user control who can see what. Pick it over Cloud Firestore when your data is simple and speed of sync matters more than complex searching. The free Spark plan includes 1 GB, and the Blaze plan bills by storage and download.","useCases":["Show who is online now","Update live scoreboards instantly","Sync simple data across devices"],"pricing":"Spark free tier (1 GB). Blaze billed by storage and download.","status":"ga","url":"https://firebase.google.com/products/realtime-database","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-sql-connect","provider":"gcp","name":"Firebase SQL Connect","category":"sql","also":["realtime-db"],"tagline":"Use a Postgres database in your app with ready-made code and live queries.","description":"Firebase SQL Connect lets your app talk to a managed Postgres, a popular database that stores data in tables, running on Cloud SQL. If your product has linked data like customers, orders and products, you describe it once and SQL Connect generates safe, ready-to-use code for Kotlin, iOS, Flutter and the web. It uses GraphQL (a way to ask for exactly the data you need) and also allows raw SQL. It is the new name for Firebase Data Connect. Pick it over Firestore when your data fits tables better. Cloud SQL costs apply, and a no-cost trial instance is available.","useCases":["Back apps with structured database tables","Run live queries on Postgres data","Give mobile apps safe, organized data access"],"pricing":"Cloud SQL costs apply; no-cost trial instance available.","status":"ga","url":"https://firebase.google.com/products/sql-connect","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-hosting","provider":"gcp","name":"Firebase Hosting","category":"hosting","also":["cdn"],"tagline":"Put your website online fast, with HTTPS and your own domain.","description":"Firebase Hosting puts your website's files on the internet and delivers them quickly to visitors around the world through a CDN (a network of servers close to your users). If you have a marketing site or a single-page app, you upload it and get secure HTTPS and your own domain. Preview channels give each change its own test link, handy for reviewing updates before they go live. Pick it over a full server when your site is mostly fixed files. It can also pass some requests to Cloud Run or Functions for extra logic. The free Spark plan includes 10 GB of storage, and Blaze is pay as you go.","useCases":["Publish marketing websites quickly","Serve single-page web apps","Create preview links for each change"],"pricing":"Spark free (10 GB storage, 360 MB/day). Blaze pay as you go.","status":"ga","url":"https://firebase.google.com/products/hosting","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-app-hosting","provider":"gcp","name":"Firebase App Hosting","category":"hosting","also":["serverless"],"tagline":"Publish Next.js and Angular apps straight from your GitHub code.","description":"Firebase App Hosting builds and publishes full apps made with Next.js or Angular, two popular web frameworks, straight from your GitHub repository. If your site builds pages on the server, such as personalized pages for each visitor, you push your code and it handles the build and the launch. It runs on Google's Cloud Build, Cloud Run and Cloud CDN, and can keep passwords and API keys safe with Secret Manager. Pick it over Firebase Hosting when your app needs a server, not just fixed files. It needs the Blaze plan, where you pay for the Cloud Run, Build and CDN you use.","useCases":["Deploy Next.js apps straight from GitHub","Host Angular web apps","Publish server-rendered websites"],"pricing":"Blaze plan. Pay for Cloud Run, Build and CDN usage.","status":"ga","url":"https://firebase.google.com/products/app-hosting","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-cloud-functions","provider":"gcp","name":"Cloud Functions for Firebase","category":"serverless","tagline":"Run your own backend code whenever something happens in your app.","description":"Cloud Functions for Firebase lets you write small pieces of backend code, in Node.js or Python, that run when something happens. If you want to send a welcome email when someone signs up, or receive a webhook (a message another service sends your app), a function does it without you managing a server. It can react to Auth, Firestore, Storage and web requests, and runs on Cloud Run functions underneath. Pick it when your app needs private logic that should not live on the user's device. It needs the Blaze plan and costs the same as Cloud Run functions.","useCases":["Receive webhooks from other services","React when database records change","Run backend code on app requests"],"pricing":"Blaze plan. Same pricing as Cloud Run functions.","status":"ga","url":"https://firebase.google.com/products/functions","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-cloud-storage","provider":"gcp","name":"Cloud Storage for Firebase","category":"object-storage","tagline":"Let users upload and download photos and files safely in your app.","description":"Cloud Storage for Firebase lets people upload and download files in your app, such as photos, documents and videos. If your app has profile pictures, you add a few lines of code and uploads work even on shaky connections, because they can resume where they stopped. Security rules decide who can see or change each file. Under the hood it uses Google Cloud Storage buckets (containers for files). Pick it over setting up storage yourself when you already use Firebase. It needs the Blaze plan with Cloud Storage prices, and the default bucket has a small no-cost quota.","useCases":["Let users upload photos","Store user documents securely","Serve media files to the app"],"pricing":"Blaze plan; Cloud Storage prices apply. Small no-cost quota in default bucket.","status":"ga","url":"https://firebase.google.com/products/storage","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-app-check","provider":"gcp","name":"Firebase App Check","category":"bots","also":["identity"],"tagline":"Make sure requests to your backend come from your real app.","description":"Firebase App Check helps make sure the requests reaching your backend come from your real app and not from copies or scripts. If bots are hammering your database or running up your bill, App Check attaches a proof of authenticity (from Play Integrity, App Attest or reCAPTCHA) to each call, and your backend turns away anything without it. You can use it to protect Firestore and Functions, and to guard calls to Gemini, Google's AI model. Pick it as an extra layer on top of sign-in, since it checks the app rather than the person. It is free for most providers.","useCases":["Block requests not from your app","Protect database and backend calls","Guard AI model calls from abuse"],"pricing":"Free for most providers.","status":"ga","url":"https://firebase.google.com/products/app-check","source":"https://firebase.google.com/products-build","gem":"Block requests that do not come from your genuine app."},{"id":"gcp-firebase-phone-number-verification","provider":"gcp","name":"Firebase Phone Number Verification","category":"identity","also":["bots"],"tagline":"Confirm a user's phone number through their carrier, with no text code.","description":"Firebase Phone Number Verification checks that someone owns a phone number by asking their mobile carrier directly, so nobody has to wait for a text message and type in a code. If your app has a sign-up screen, people can tap once to confirm their number and move on. It also helps cut down on text-message (SMS) fraud, where bad actors abuse sign-up codes. It became generally available in May 2026 on Android, with more than 10 carriers. You pay as you go, and a billing account is required.","useCases":["Onboard users without SMS codes","Verify phone numbers without texted codes","Cut fake signups and SMS fraud"],"pricing":"Pay as you go; billing account required.","status":"ga","url":"https://firebase.google.com/products/phone-number-verification","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-ai-logic","provider":"gcp","name":"Firebase AI Logic","category":"ai-models","tagline":"Add Gemini AI chat and image features straight into your mobile or web app.","description":"Firebase AI Logic lets your mobile or web app talk to Gemini, Google's AI model, using ready-made toolkits for Swift, Kotlin, JavaScript, Dart and Unity. If you want an in-app chat helper or a feature that describes photos users upload, this is a direct way to add it. It includes App Check, which blocks traffic from apps that aren't yours, plus per-user rate limits so one person can't run up your bill. The Gemini Developer API has a free tier, while the Vertex AI route is pay as you go.","useCases":["Add in-app chat powered by Gemini","Analyze images inside mobile apps","Mix on-device and cloud AI features"],"pricing":"Gemini Developer API has a free tier; Vertex path is pay as you go.","status":"ga","url":"https://firebase.google.com/products/firebase-ai-logic","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-genkit","provider":"gcp","name":"Genkit","category":"ai-platform","also":["devtools","agents"],"tagline":"Free toolkit for building AI features, from chatbots to AI agents.","description":"Genkit is a free, open source toolkit (its code is public) for building AI features in TypeScript, Go or Python. If you want a chatbot that answers questions from your own documents, a pattern known as RAG, it gives you one consistent way to connect AI models, tools and your data. A local developer screen lets you test things as you build. You can deploy to Cloud Functions, Cloud Run (Google's services for running your code) or anywhere else. You only pay for model calls and hosting.","useCases":["Build AI agents and multi-step flows","Connect AI models to your own data","Prototype AI features across different models"],"pricing":"Free open source. Pay for model calls and hosting.","status":"ga","url":"https://firebase.google.com/products/genkit","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-extensions","provider":"gcp","name":"Firebase Extensions","category":"devtools","also":["serverless"],"tagline":"Ready-made backend add-ons for common jobs (ending March 31, 2027).","description":"Firebase Extensions were ready-made add-ons that handled common behind-the-scenes jobs for your app, such as resizing images or copying data from Firestore, Firebase's database, into BigQuery, Google's tool for analyzing large amounts of data. If your app lets people upload photos, an extension could shrink them automatically without you writing that code. You paid only for the underlying functions it ran. Extensions are deprecated and being replaced by function kits, and the service ends March 31, 2027.","useCases":["Keep existing add-ons running until 2027","Plan a move to function kits"],"pricing":"Pay for underlying functions.","status":"deprecated","url":"https://firebase.google.com/docs/extensions","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-studio","provider":"gcp","name":"Firebase Studio","category":"devtools","also":["ai-platform"],"tagline":"Online coding workspace with AI prototyping, shutting down March 22, 2027.","description":"Firebase Studio is an online coding workspace, sometimes called a cloud IDE, where you can build and test app ideas with AI help. If you sketched a prototype there, you will need to move that project elsewhere, and Google recommends Google AI Studio or Antigravity. Existing workspaces stay free during the sunset period. New workspaces were disabled on June 22, 2026, and the service shuts down March 22, 2027.","useCases":["Keep existing workspaces running until shutdown","Move projects to Google AI Studio"],"pricing":"Free during sunset period.","status":"deprecated","url":"https://firebase.google.com/docs/studio","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-ml","provider":"gcp","name":"Firebase ML","category":"ai-platform","tagline":"Run AI models on phones and use image analysis; shutting down June 15, 2027.","description":"Firebase ML hosted small AI models that run right on a person's phone, using TensorFlow Lite (a format for compact models), and passed image requests to Cloud Vision, Google's image analysis service. If your app recognized objects in photos, it could have used this. Google now points to Cloud Storage, its file storage service, for hosting model files and to Firebase AI Logic for AI features. Projects that did not already use it cannot start. Firebase ML is deprecated and shuts down June 15, 2027.","useCases":["Keep existing projects running for now","Move models to Cloud Storage"],"pricing":"Existing projects only.","status":"closed","url":"https://firebase.google.com/docs/ml","source":"https://firebase.google.com/docs/ml"},{"id":"gcp-firebase-cloud-messaging","provider":"gcp","name":"Firebase Cloud Messaging","category":"messaging","tagline":"Free push notifications for Android, iOS and web apps.","description":"Firebase Cloud Messaging sends push notifications, the small alerts that pop up on a phone, to your users' devices. If you want to tell people their order shipped, or broadcast news to everyone who follows a topic, it handles the delivery. It can also send silent messages that update data in your app without showing anything. It is the delivery layer used by most Android apps, and it connects with analytics so you can see how messages perform. It is free, with no limits.","useCases":["Send push notifications to phones","Broadcast messages to topic groups","Send quiet background data updates"],"pricing":"Free, unlimited.","status":"ga","url":"https://firebase.google.com/products/cloud-messaging","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-in-app-messaging","provider":"gcp","name":"Firebase In-App Messaging","category":"messaging","also":["analytics"],"tagline":"Show cards, banners and pop-ups inside your app to the right users.","description":"Firebase In-App Messaging lets you show cards, banners and pop-up windows inside your app while people are using it. If new users tend to stall after sign-up, you can show a nudge that guides them to the next step, or promote a sale. You choose who sees what based on actions people take and audiences built in Google Analytics. Unlike push notifications, these appear only when someone already has your app open. It is free.","useCases":["Show onboarding tips inside the app","Display promotions to chosen users"],"pricing":"Free.","status":"ga","url":"https://firebase.google.com/products/in-app-messaging","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-remote-config","provider":"gcp","name":"Firebase Remote Config","category":"feature-flags","tagline":"Change how your app looks and behaves without releasing a new version.","description":"Firebase Remote Config lets you change settings in your app from a dashboard, so you don't have to publish a new version and wait for app store approval. If you want to turn on a new feature for 10 percent of users first, you set a condition and roll it out gradually. It also hooks into A/B tests, and it can serve the prompt templates used by Firebase AI Logic, so you can adjust AI wording on the fly. It is free.","useCases":["Turn features on without releases","Roll out changes gradually","Tune AI prompts while running"],"pricing":"Free.","status":"ga","url":"https://firebase.google.com/products/remote-config","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-ab-testing","provider":"gcp","name":"Firebase A/B Testing","category":"feature-flags","also":["analytics"],"tagline":"Test versions of a feature, message or AI model and pick the winner.","description":"Firebase A/B Testing helps you try two or more versions of something and see which works better. If you're unsure whether a friendlier notification message gets more taps, you can send each version to a slice of users and compare the results. It builds on Remote Config and Cloud Messaging, uses Google Analytics for the statistics, and can roll the winner out to everyone. You can test screen layouts, notification wording and even different AI models. It is free.","useCases":["Test two app screen designs","Compare notification wording options","Compare AI model versions"],"pricing":"Free.","status":"ga","url":"https://firebase.google.com/products/ab-testing","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-crashlytics","provider":"gcp","name":"Firebase Crashlytics","category":"observability","tagline":"See when and why your app crashes, with AI suggestions for fixes.","description":"Firebase Crashlytics reports crashes and smaller errors from your live app as they happen. If users start leaving bad reviews because the app keeps closing, it groups similar crashes together and shows the trail of code and the steps leading up to each one. Gemini, Google's AI model, adds suggestions for what to try. It works with iOS, Android, Flutter and Unity apps, and it can alert you when errors trend upward. It is free.","useCases":["Find and group app crashes","Track overall app health","Alert when errors suddenly rise"],"pricing":"Free.","status":"ga","url":"https://firebase.google.com/products/crashlytics","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-performance-monitoring","provider":"gcp","name":"Firebase Performance Monitoring","category":"observability","tagline":"Measure how fast your app starts, draws screens and talks to the internet.","description":"Firebase Performance Monitoring measures how fast your app feels on real people's phones. If users complain the app is slow to open, you can see startup times and find which screens or network requests drag. It collects traces, which are timed recordings of tasks in your app, plus network request data from real devices rather than just your own test phone. It is free.","useCases":["Find why the app starts slowly","Track slow network calls on real phones","Measure how fast screens draw"],"pricing":"Free.","status":"ga","url":"https://firebase.google.com/products/performance","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-test-lab","provider":"gcp","name":"Firebase Test Lab","category":"devtools","tagline":"Try your app on many real and virtual phones before you release it.","description":"Firebase Test Lab runs your app on lots of different phones and tablets, some real and some virtual, all hosted in Google's data centers. If you are about to ship an update, you can check that it still works across a range of devices without buying any of them. It supports instrumentation tests (checks your developers script), Robo tests (an automatic tour through your app) and game loop tests. Pick it when you want to catch device-specific bugs before your users do. There is a free daily quota on the Spark plan, and the Blaze plan charges per device minute.","useCases":["Test the app on many devices","Catch bugs before each release","Run automated tests on virtual phones"],"pricing":"Free daily quota on Spark; Blaze per device minute.","status":"ga","url":"https://firebase.google.com/products/test-lab","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-app-distribution","provider":"gcp","name":"Firebase App Distribution","category":"devtools","tagline":"Send test versions of your app to testers by email.","description":"Firebase App Distribution sends early versions of your iOS and Android apps to testers by email. If you are building a mobile app and want a few friends, customers or a QA team to try it before launch, you invite them and they get the build, with links to send feedback and see crashes. It is free, which makes it a good fit for small teams running a beta. Pick it when you want a simple, no-cost way to get early builds into people's hands.","useCases":["Send test builds to beta users by email","Collect tester feedback and crash reports","Share QA builds with your team"],"pricing":"Free.","status":"ga","url":"https://firebase.google.com/products/app-distribution","source":"https://firebase.google.com/products-build"},{"id":"gcp-firebase-google-analytics","provider":"gcp","name":"Google Analytics for Firebase","category":"analytics","tagline":"Free app analytics that tracks what people do and groups them into audiences.","description":"Google Analytics for Firebase records the actions people take in your app, such as opening a screen or buying something, and groups users into audiences. If you want to know which people finished signing up and then message just those who did not, this is where you would start. Those audiences also feed Firebase's Messaging and Remote Config tools, and you can export the data to BigQuery, Google's tool for analyzing large piles of data. It is free with unlimited events, so you will not pay more as your app grows.","useCases":["Track which events users trigger in the app","Build user groups for messaging and remote config","Export app event data to BigQuery"],"pricing":"Free, unlimited events.","status":"ga","url":"https://firebase.google.com/products/analytics","source":"https://firebase.google.com/products-build"},{"id":"akamai-compute-instances","provider":"akamai","name":"Akamai Cloud Compute Instances","category":"vms","tagline":"Rent Linux servers by the hour or month, from about $5 a month.","description":"Akamai Cloud Compute Instances are Linux virtual machines (servers you rent instead of buying) that come in shared, dedicated, premium and high-memory plans. If you need a place to host your website or API, or a cheap server for testing, you can start one in minutes. They are sold under the Akamai Cloud brand since Akamai acquired Linode in 2022. Pick them over bigger clouds when you want flat monthly pricing and plenty of bundled data transfer. Shared plans start around $5 a month, and you can pay hourly or monthly.","useCases":["Host websites and APIs on virtual servers","Spin up dev and test servers","Run your own databases on servers"],"pricing":"Hourly or monthly. Shared plans from about $5/month.","status":"ga","url":"https://www.akamai.com/cloud/compute","source":"https://www.akamai.com/products"},{"id":"akamai-gpu-compute","provider":"akamai","name":"Akamai Cloud GPU and Accelerated Compute","category":"gpu","tagline":"Rent servers with graphics chips for AI, video and rendering work.","description":"Akamai Cloud GPU and Accelerated Compute gives you servers with GPUs, which are chips that handle AI and graphics work much faster than ordinary processors. If your app runs a chatbot, converts video or renders images, you would use these servers to do the heavy lifting. In 2026 the lineup expanded to include NVIDIA RTX PRO 6000 Blackwell GPUs. Pick it when you want AI answers to come back quickly because the servers sit close to your users. Pricing is hourly per GPU plan.","useCases":["Run AI models for fast user responses","Render graphics and images at speed","Convert video into different formats"],"pricing":"Hourly per GPU plan. See pricing page.","status":"ga","url":"https://www.akamai.com/cloud/compute/gpu","source":"https://www.akamai.com/products"},{"id":"akamai-inference","provider":"akamai","name":"Akamai Inference","category":"ai-models","tagline":"Run AI models and serve answers to users from Akamai's GPU servers.","description":"Akamai Inference is a managed service for putting an AI model online so your app can send it questions and get answers back (that step is called inference). If you want to add a chat feature or an AI helper to your product using an open model, you deploy it here instead of setting up the hardware yourself. It runs on GPU servers placed near end users and is part of Akamai Cloud for AI. Pick it when quick responses matter. Pricing is usage-based, and you contact sales for committed plans.","useCases":["Serve open AI models to users","Add fast AI features near users","Power backends for AI agents"],"pricing":"Usage-based. Contact sales for committed plans.","status":"ga","url":"https://www.akamai.com/products/akamai-inference","source":"https://www.akamai.com/products"},{"id":"akamai-lke","provider":"akamai","name":"Linode Kubernetes Engine (LKE)","category":"containers","tagline":"Managed Kubernetes on Akamai Cloud, with a free standard control plane.","description":"Linode Kubernetes Engine (LKE) is a managed way to run Kubernetes, a system that starts, stops and keeps watch over your containers (apps packaged to run anywhere). If your product is split into several small services, or you need to run AI jobs on GPU servers, LKE keeps them all running for you. You only pay for the servers, called nodes, because the standard control plane, the part that does the managing, is free. An enterprise tier adds a high-availability control plane with an SLA, which is a written uptime promise.","useCases":["Run microservices in containers","Run AI workloads on GPU nodes","Run build and test jobs"],"pricing":"Pay for nodes. Standard control plane is free.","status":"ga","url":"https://www.akamai.com/cloud/kubernetes","source":"https://www.akamai.com/products"},{"id":"akamai-app-platform","provider":"akamai","name":"App Platform for LKE","category":"containers","also":["devtools"],"tagline":"A ready-made setup for running apps on Kubernetes, with tools pre-connected.","description":"App Platform for LKE is a pre-built layer on top of Linode Kubernetes Engine that bundles more than 30 open source tools from the CNCF ecosystem. These cover CI/CD (automatic testing and releasing), ingress (letting traffic in), secrets, GitOps (deploying from your code repository) and observability (watching how things run). If your team wants its own internal platform for shipping apps, this saves you from wiring those tools together by hand. There is no extra fee, and you only pay for the LKE resources you use.","useCases":["Give teams a ready internal developer platform","Host apps for many separate customers","Deploy apps with GitOps workflows"],"pricing":"No extra fee. Pay for LKE resources.","status":"ga","url":"https://www.akamai.com/products/app-platform","source":"https://www.akamai.com/products"},{"id":"akamai-functions","provider":"akamai","name":"Akamai Functions","category":"serverless","also":["edge-compute"],"tagline":"Run small bits of code close to your users, built on WebAssembly.","description":"Akamai Functions is a serverless platform, meaning you upload small pieces of code and Akamai runs them without you managing any servers. The code runs as WebAssembly (Wasm), a compact format that starts quickly, across Akamai's distributed network. If you need an API that answers fast for people around the world, or logic that decides where each request should go, this fits. It is powered by Akamai's 2025 acquisition of Fermyon. You can write in Rust, Go, JavaScript, Python and other languages that compile to Wasm. Pricing is usage-based.","useCases":["Run APIs close to users","Handle AI requests at the edge","Route requests with custom logic"],"pricing":"Usage-based. See Akamai pricing page.","status":"ga","url":"https://www.akamai.com/products/akamai-functions","source":"https://www.akamai.com/products"},{"id":"akamai-agent-sandbox","provider":"akamai","name":"Akamai Agent Sandbox","category":"agents","tagline":"Safe, walled-off spaces to run AI agents and code you don't fully trust.","description":"Akamai Agent Sandbox gives you isolated spaces for running AI agents, the kind of AI that takes actions for you, along with code you cannot fully trust. If your product lets an agent write and run code or call tools, a sandbox keeps any mistakes away from the rest of your system. It is listed under Akamai's serverless products. Details are thin so far, so treat how mature it is as unconfirmed. It is currently in preview, and pricing is by contacting sales.","useCases":["Run code from AI agents safely","Isolate tools that agents call"],"pricing":"Contact sales.","status":"preview","url":"https://www.akamai.com/products","source":"https://www.akamai.com/products"},{"id":"akamai-object-storage","provider":"akamai","name":"Akamai Object Storage","category":"object-storage","tagline":"Store files and images with S3-compatible tools, with transfer bundled in.","description":"Akamai Object Storage, formerly Linode Object Storage, is a place to keep files such as images, backups and datasets. It is S3-compatible, which means it works with the many tools built for Amazon S3. If your app lets people upload photos, you can keep them here and serve them back later. The pricing is flat and bundles storage with outbound data transfer, so bills are easier to predict. It starts at $5 a month, including 250 GB of storage and 1 TB of transfer.","useCases":["Store backups in the cloud","Serve static assets for websites","Hold datasets for AI projects"],"pricing":"From $5/month including 250 GB and 1 TB transfer.","status":"ga","url":"https://www.akamai.com/cloud/storage/object-storage","source":"https://www.akamai.com/products"},{"id":"akamai-block-storage","provider":"akamai","name":"Akamai Block Storage","category":"block-storage","tagline":"Add extra disk space to your Akamai servers, at $0.10 per GB a month.","description":"Akamai Block Storage, formerly Linode Block Storage, gives you extra SSD disk space that you attach to your Akamai Cloud servers. If your database is outgrowing the disk that came with your server, you can add a volume and keep going without moving to a bigger plan. These volumes connect over the network and can serve as database storage or as more room for any instance. Pick it when you just need more space for a server you already run. It costs $0.10 per GB-month.","useCases":["Attach disks for database storage","Add extra disk space to servers"],"pricing":"$0.10/GB-month.","status":"ga","url":"https://www.akamai.com/cloud/storage/block-storage","source":"https://www.akamai.com/products"},{"id":"akamai-backups","provider":"akamai","name":"Akamai Cloud Backups","category":"block-storage","tagline":"Automatic copies of your Akamai server, so you can restore it if things break.","description":"Akamai Cloud Backups takes saved copies of your server (called an instance) on a daily, weekly or every-other-week schedule, and you can also take one by hand whenever you like. If you are about to push a risky update to your app, take a manual snapshot first, and if the update goes wrong you can roll back to how things were. It is also your safety net if a server is lost or damaged. It is an add-on you pay for each month per server, with the price based on the size of your plan.","useCases":["Recover from disaster with snapshots","Roll back after a bad deploy"],"pricing":"Per-instance monthly add-on based on plan size.","status":"ga","url":"https://www.akamai.com/products","source":"https://www.akamai.com/products"},{"id":"akamai-managed-databases","provider":"akamai","name":"Akamai Managed Databases","category":"sql","tagline":"A ready-to-run MySQL or PostgreSQL database that Akamai looks after for you.","description":"Akamai Managed Databases gives you a database for MySQL or PostgreSQL (two popular ways to store app data) without having to set it up or maintain it yourself. If your app needs a place to keep user accounts, orders or posts, this is where they would live. Akamai handles automated backups, and you can choose options that keep the database running if one machine fails. Pick it over running your own database when you would rather build your product than babysit servers. Billing is hourly by size, with a single-node setup starting around $15 a month.","useCases":["Run MySQL or PostgreSQL databases","Keep databases running with failover","Power read-heavy app services"],"pricing":"Hourly by node size. Standalone from about $15/month.","status":"ga","url":"https://www.akamai.com/cloud/databases","source":"https://www.akamai.com/products"},{"id":"akamai-nodebalancers","provider":"akamai","name":"NodeBalancers","category":"load-balancing","tagline":"Spread visitors across several servers so your site stays up under load.","description":"NodeBalancers is a load balancer, a traffic director that sends each visitor to one of several servers. If your app gets popular and one server can no longer keep up, you add more servers and put a NodeBalancer in front of them. It checks that each server is healthy and stops sending people to any that are not. It can also handle the security certificate for HTTPS (TLS), so your servers do less work. Choose it when you want more capacity and fewer outages on Akamai Cloud. It starts at $10 a month per balancer.","useCases":["Spread traffic across web servers","Keep sites up during failures"],"pricing":"From $10/month per balancer.","status":"ga","url":"https://www.akamai.com/cloud/networking","source":"https://www.akamai.com/products"},{"id":"akamai-cloud-firewall","provider":"akamai","name":"Akamai Cloud Firewall","category":"networking","tagline":"A free firewall that decides who can reach your Akamai servers.","description":"Akamai Cloud Firewall lets you set rules for what traffic can reach your servers and load balancers, and it blocks everything else. For example, you can allow only your own office to log in to a server remotely (over SSH, a standard remote login method), which keeps strangers out. You can also separate parts of your app, so a public web server cannot freely talk to your database. There is nothing to install on your servers, because the rules are applied from the outside. It is free, so there is little reason to skip it if you run things on Akamai Cloud.","useCases":["Block SSH access from outside","Separate app tiers from each other"],"pricing":"Free.","status":"ga","url":"https://www.akamai.com/cloud/networking","source":"https://www.akamai.com/products"},{"id":"akamai-dns-manager","provider":"akamai","name":"DNS Manager","category":"dns","tagline":"Free DNS hosting that points your domain name to your Akamai servers.","description":"DNS Manager is the address book that tells the internet which server to visit when someone types your domain name. If you have launched a small site or app on Akamai Cloud and bought a domain elsewhere, you can manage its records here, such as which name leads to which server. Because it lives in the same place as your servers, you can change settings without juggling another account. It is free, which makes it an easy choice for small sites. You manage domains used with your Akamai instances.","useCases":["Host DNS records for a small site","Route a domain to your cloud servers","Manage domain settings for your instances"],"pricing":"Free.","status":"ga","url":"https://www.akamai.com/cloud/networking","source":"https://www.akamai.com/products"},{"id":"akamai-vpc","provider":"akamai","name":"VPC and Private Networking","category":"networking","tagline":"Private networks so your Akamai servers can talk without using the public internet.","description":"VPC and Private Networking lets you create a closed-off network (a VPC, or virtual private cloud) where your servers and clusters can talk to each other privately. If your app has a web server and a separate database, you can keep the database reachable only from inside this private network, so it is never exposed to the open internet. This is a good way to organize a multi-part app and keep internal traffic away from outsiders. Pick it when you want internal traffic kept private. It costs nothing within a single region.","useCases":["Connect servers over private networks","Keep internal service traffic private","Build multi-tier network layouts"],"pricing":"Free within a region.","status":"ga","url":"https://www.akamai.com/cloud/networking","source":"https://www.akamai.com/products"},{"id":"akamai-ion","provider":"akamai","name":"Akamai Ion","category":"cdn","tagline":"Makes busy, changing websites and apps load faster for visitors worldwide.","description":"Akamai Ion is a content delivery network (CDN, a worldwide set of servers that sits close to your visitors) built for sites whose pages change often. If you run an online store where prices, carts and pages differ from person to person, Ion speeds up those pages instead of only the unchanging files. It runs on Akamai's very large global network, which is a big reason to pick it for large or high-traffic sites. Pricing is not listed, so you will need to contact sales.","useCases":["Speed up ecommerce product pages","Deliver large dynamic sites faster"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/ion","source":"https://www.akamai.com/products"},{"id":"akamai-api-acceleration","provider":"akamai","name":"API Acceleration","category":"cdn","tagline":"Speeds up and steadies the data calls your app and partners make.","description":"API Acceleration makes API calls (the requests your app sends to fetch or save data) faster and more dependable. If you have a mobile app that talks to your backend all day, or partners who send you data, this service improves how those requests travel across the Akamai network. It picks better routes, stores repeat answers, and reuses connections so each call does not start from scratch. Choose it when slow or flaky data requests are hurting your users. Pricing is not listed, so you will need to contact sales.","useCases":["Speed up calls from mobile apps","Make partner API traffic more reliable"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/api-acceleration","source":"https://www.akamai.com/products"},{"id":"akamai-adaptive-media-delivery","provider":"akamai","name":"Adaptive Media Delivery","category":"video","tagline":"Smooth video streaming that adjusts quality to each viewer's connection.","description":"Adaptive Media Delivery is a content delivery network (CDN, a set of servers close to viewers) set up for streaming video to very large audiences. It supports HLS and DASH, two common ways of streaming where video quality shifts up or down to match each viewer's connection. If you run a streaming service or broadcast live sports, this helps everyone watch without constant buffering. Pick it over a general-purpose CDN when video is your main business. Pricing is not listed, so you will need to contact sales.","useCases":["Stream video to large audiences","Deliver live sports to viewers"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/adaptive-media-delivery","source":"https://www.akamai.com/products"},{"id":"akamai-download-delivery","provider":"akamai","name":"Download Delivery","category":"cdn","tagline":"Gets big files like game patches and installers to users quickly.","description":"Download Delivery is a content delivery network (CDN, a set of servers spread around the world) tuned for very large files. If you ship a game, an app installer or regular software updates, thousands of people may try to download them at once, and this service helps those downloads arrive reliably instead of crawling. Pick it over a general-purpose CDN when your files are big and your audience is large. Pricing is not listed, so you will need to contact sales.","useCases":["Deliver game patches quickly","Distribute software updates to users"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/download-delivery","source":"https://www.akamai.com/products"},{"id":"akamai-media-services-live","provider":"akamai","name":"Media Services Live","category":"video","tagline":"Run a live video broadcast from camera feed to viewers' screens.","description":"Media Services Live is a managed service that handles live video from start to finish. If you are hosting a live event or running a channel that streams around the clock, you send in your video feed, and the service takes care of encoding it (converting it into formats that work on different devices) and delivering it to viewers. That means you do not have to piece together separate tools for each step. Pick it when you want one provider to cover the whole live workflow. Pricing is not listed, so you will need to contact sales.","useCases":["Stream live events to viewers","Run linear channels around the clock","Ingest and encode live video"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/media-services-live","source":"https://www.akamai.com/products"},{"id":"akamai-image-video-manager","provider":"akamai","name":"Image & Video Manager","category":"images","tagline":"Automatically shrinks and resizes your images and videos for each visitor.","description":"Image & Video Manager adjusts your pictures and videos on the fly so each visitor gets a version that fits their device and connection. If your site shows large photos, a phone on a slow network gets a smaller, lighter file while a big desktop screen gets a sharper one. It resizes, converts and compresses automatically, which saves bandwidth (the data your site sends out) and helps pages load faster. Pick it when you do not want to prepare many versions of every file by hand. Pricing is not listed, so you will need to contact sales.","useCases":["Resize images for each device","Cut bandwidth on images and video","Serve responsive images automatically"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/image-and-video-manager","source":"https://www.akamai.com/products"},{"id":"akamai-edgeworkers","provider":"akamai","name":"EdgeWorkers","category":"edge-compute","tagline":"Run small JavaScript snippets close to your visitors on Akamai's network.","description":"EdgeWorkers lets you run small pieces of JavaScript on Akamai's network, on servers close to your visitors, before a request ever reaches your own server. If you run a store and want to send visitors from different countries to different pages, or split traffic between two versions of a landing page for an A/B test, that logic can live here. It also handles redirects and changes to headers (small bits of info sent along with each web request). Akamai positions it as a companion to Akamai Functions. Pricing is through Akamai sales.","useCases":["Run custom logic at the edge","Personalize pages for each visitor","Redirect visitors and change headers"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/edgeworkers","source":"https://www.akamai.com/products"},{"id":"akamai-edgekv","provider":"akamai","name":"EdgeKV","category":"kv-cache","tagline":"A fast lookup table that Akamai keeps copies of near your visitors.","description":"EdgeKV is a simple database that stores values under a name (a key-value store) and keeps copies across Akamai's global network so lookups are quick. It is built to work with EdgeWorkers. If you want to turn a new feature on for some visitors, or keep a list of old page addresses and where each should redirect, you can store that data here and read it from your edge code. Pick it when your edge logic needs data close by. Pricing is through Akamai sales.","useCases":["Store feature toggles close to users","Keep redirect maps for fast lookups","Serve shared data to edge code"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/edgekv","source":"https://www.akamai.com/products"},{"id":"akamai-cloudlets","provider":"akamai","name":"Cloudlets","category":"edge-compute","tagline":"Ready-made apps for redirects, visitor queues and traffic rules.","description":"Cloudlets are small, ready-made apps for common traffic rules, managed from Akamai's control center instead of written as code. If you are launching a popular product drop and expect more visitors than your site can handle, the waiting room option lines people up and lets them in gradually. The same approach covers redirects and forwarding rules that send visitors to the right place. Pick Cloudlets over writing your own edge code when you want a point-and-click setup. Pricing is through Akamai sales.","useCases":["Hold visitors in a waiting room","Set up redirect and forwarding rules","Prioritize visitors by audience"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/cloudlets","source":"https://www.akamai.com/products"},{"id":"akamai-cloud-wrapper","provider":"akamai","name":"Cloud Wrapper","category":"cdn","tagline":"A cache that eases the load and data fees on your cloud storage.","description":"Cloud Wrapper adds a cache run by Akamai that sits between your cloud storage and Akamai's delivery network. It holds copies of your content nearby, so your storage gets asked for files less often. If you stream video or serve large media files from a cloud storage bucket, you can cut egress (the fee clouds charge when data leaves them) and put less strain on that storage. Pick it when your files already live in a cloud and data-out bills are the pain. Pricing is through Akamai sales.","useCases":["Cut egress costs from your cloud","Serve large media with less origin load","Cache files next to cloud storage"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/cloud-wrapper","source":"https://www.akamai.com/products"},{"id":"akamai-gtm","provider":"akamai","name":"Global Traffic Management","category":"load-balancing","tagline":"Sends visitors to the best data center and switches over if one fails.","description":"Global Traffic Management is a traffic director that works through DNS, the internet's address book that turns your domain name into a server location. It sends each visitor to the best data center based on health, geography and how busy each one is. If your app runs in two regions and you want people to land on the nearest one, or move over automatically when one goes down, this does the job. Pick it when you run in several places and want failover (switching to a backup) handled early. Pricing is through Akamai sales.","useCases":["Fail over to a backup region","Send users to the nearest data center","Balance traffic by health and load"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/global-traffic-management","source":"https://www.akamai.com/products"},{"id":"akamai-data-stream","provider":"akamai","name":"DataStream","category":"observability","tagline":"Sends your Akamai traffic logs to your own tools within moments.","description":"DataStream sends logs about how your content is delivered and how it performs from Akamai's network to your own storage or analytics tools, in near real time. If your site suddenly slows down or starts returning errors, you can look at what was happening right then instead of waiting for a report. It also helps you understand traffic patterns over time. Pick it when you want the raw delivery data in the tools your team already uses. Pricing is through Akamai sales.","useCases":["Stream CDN logs into your analytics","Find delivery problems faster","Send performance logs to your storage"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/datastream-2","source":"https://www.akamai.com/products"},{"id":"akamai-mpulse","provider":"akamai","name":"mPulse","category":"observability","tagline":"See how fast your site is for real visitors and how it affects results.","description":"mPulse is real user monitoring, which means it measures how fast your website or mobile app feels to the actual people using it. It then ties that speed data to business numbers. If you want to know whether a slow page is costing you sign-ups or purchases, you can track page speed and compare it with conversions. Pick it over a lab-style speed test when you care about what real visitors on real devices experience. Pricing is through Akamai sales.","useCases":["Track real page speed for users","Connect speed to conversion results","Measure web and mobile performance"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/mpulse-real-user-monitoring","source":"https://www.akamai.com/products"},{"id":"akamai-cloudtest","provider":"akamai","name":"CloudTest","category":"observability","tagline":"Test how your site or API holds up under heavy traffic before launch.","description":"CloudTest is a load testing platform that simulates large amounts of traffic against your website or API (the connection point other software uses to talk to your app). If you are about to launch a big campaign or sale, you can rehearse the rush first and see where things slow down or break. You can also rerun the same tests after changes to catch new slowdowns. Pick it when you want to find limits before real visitors do. Pricing is through Akamai sales.","useCases":["Simulate traffic before a big launch","Test APIs under heavy load","Run regression load tests"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/cloudtest","source":"https://www.akamai.com/products"},{"id":"akamai-edge-dns","provider":"akamai","name":"Edge DNS","category":"dns","tagline":"Hosts your domain's address records on Akamai's global network.","description":"Edge DNS is a managed service that answers the question \"where does this domain live?\" for everyone who visits your site. It is built for high availability, so it keeps answering even when parts of the internet struggle, and it supports DNSSEC, a security feature that helps stop DNS answers from being faked. If your brand cannot afford downtime, you can use it as your main DNS. Akamai positions it as the DNS layer of its security stack. Pricing is through Akamai sales.","useCases":["Host your main DNS records","Keep DNS online for large sites","Secure DNS with DNSSEC"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/edge-dns","source":"https://www.akamai.com/products"},{"id":"akamai-dns-posture-management","provider":"akamai","name":"DNS Posture Management","category":"dns","also":["threat-intel"],"tagline":"Scans your domain settings for risky records that could be hijacked.","description":"DNS Posture Management scans your DNS setup, the records that point your domain and subdomains to the right places, and flags mistakes and takeover risks. A common one is a dangling CNAME, a record that still points to a service you stopped using, which someone else could claim. If you have launched many test sites and campaign pages over the years, this helps you find forgotten subdomains before attackers do. Pick it for regular DNS cleanup and checks. Pricing is through Akamai sales.","useCases":["Find risky dangling DNS records","Check for subdomain takeover risk","Clean up misconfigured DNS settings"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products","source":"https://www.akamai.com/products"},{"id":"akamai-aap","provider":"akamai","name":"App & API Protector","category":"waf","tagline":"One shield against hacks, bots, traffic floods and API abuse.","description":"App & API Protector is Akamai's main protection suite for websites and APIs (the connection points other software uses to talk to your app). It combines a WAF, a filter that blocks malicious web requests, with DDoS defense against traffic floods, bot defense and API discovery. If your public API or checkout page is a target, it can block common attacks such as those on the OWASP list. It replaces the earlier Kona Site Defender line, so choose it for new setups. Pricing is through Akamai sales.","useCases":["Block OWASP-style web attacks","Protect public APIs from abuse","Stop DDoS and bot traffic"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/app-and-api-protector","source":"https://www.akamai.com/products"},{"id":"akamai-api-security","provider":"akamai","name":"API Security","category":"waf","tagline":"Finds hidden APIs and spots misuse of the ones you run.","description":"API Security helps you see and protect the APIs your business runs on. APIs are the connection points that let apps and services talk to each other. It finds shadow APIs, meaning ones nobody on your team realized were live, flags risky behavior and detects abuse. If your product has grown fast and you are not sure which endpoints are exposed, you can start with a full inventory. Pick it when you want visibility into APIs and not only a filter in front of them. Pricing is through Akamai sales.","useCases":["Discover APIs you did not know about","Flag risky API behavior","Detect API abuse at runtime"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/api-security","source":"https://www.akamai.com/products"},{"id":"akamai-client-side-protection","provider":"akamai","name":"Client-Side Protection & Compliance","category":"waf","tagline":"Spots shady scripts running in your visitors' browsers, especially on payment pages.","description":"Client-Side Protection & Compliance watches the scripts that run inside your visitors' web browsers, such as chat widgets, analytics tags and ad code. If your store takes card payments, a tampered script can quietly copy card details as people type them, an attack often called Magecart. This tool detects that kind of skimming. It also helps you meet the script rules in PCI DSS 4.0 (the card-payment security standard), specifically requirement 6.4.3. Pick it when your checkout page relies on third-party code you don't fully control. Pricing is through Akamai sales.","useCases":["Watch scripts on payment pages","Catch card-skimming scripts in browsers","Meet PCI script rules"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/client-side-protection-and-compliance","source":"https://www.akamai.com/products"},{"id":"akamai-firewall-for-ai","provider":"akamai","name":"Firewall for AI","category":"ai-gateway","tagline":"Guardrails that check what goes into and out of your AI features.","description":"Firewall for AI sits in front of your AI endpoints and screens the prompts people send and the answers that come back. If you ship a chatbot or an AI assistant in your app, someone will eventually try to trick it with a prompt injection, which is a message written to make the AI ignore its instructions. This tool helps catch those attempts and also helps stop private data from leaking out in responses. Pick it when your AI feature is public and you want protection that runs while it's live. Pricing is through Akamai sales.","useCases":["Screen prompts sent to AI models","Block prompt injection attempts","Stop sensitive data leaking via AI"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products","source":"https://www.akamai.com/products"},{"id":"akamai-bot-manager","provider":"akamai","name":"Bot Manager","category":"bots","tagline":"Tells good bots from bad ones and decides what each is allowed to do.","description":"Bot Manager sorts the automated traffic hitting your site into types, then lets you choose what happens to each: allow it, slow it down or block it. If your login page gets hammered by scripts trying stolen passwords, which is called credential stuffing, or competitors copy your listings with scrapers, this is the tool for it. Unlike a basic firewall rule that blocks one address, it is built to recognize bot behavior. It also offers advanced controls for tougher cases. Pricing is through Akamai sales.","useCases":["Stop scrapers and unwanted bots","Block credential stuffing attacks","Classify and act on bot traffic"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/bot-manager","source":"https://www.akamai.com/products"},{"id":"akamai-account-protector","provider":"akamai","name":"Account Protector","category":"bots","tagline":"Catches hijacked logins and fake signups before they do damage.","description":"Account Protector looks at every login and signup attempt and gives it a risk score based on how the person behaves and on patterns seen across other users. If your app has user accounts, this helps you spot account takeover, where someone logs in with a stolen password, and fake signups created by scripts. Choose it over a plain bot blocker when the problem is specifically about accounts, since it judges each login or signup attempt rather than all traffic. Pricing is through Akamai sales.","useCases":["Spot account takeover attempts","Catch fake account signups","Score logins using user behavior"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/account-protector","source":"https://www.akamai.com/products"},{"id":"akamai-content-protector","provider":"akamai","name":"Content Protector","category":"bots","tagline":"Stops scrapers from copying your prices, content or stock levels.","description":"Content Protector is built to catch bots that scrape your site. If you run an online shop, a travel site or a marketplace, rivals may use scripts to copy your prices, product pages or inventory, then undercut you or republish your work. This tool uses detection tuned to exactly that kind of content and pricing abuse. Pick it over a general bot tool when scraping is your main worry and you want checks focused on it. Pricing is through Akamai sales.","useCases":["Stop competitors scraping your prices","Protect content and inventory from theft","Block scrapers targeting pricing data"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/content-protector","source":"https://www.akamai.com/products"},{"id":"akamai-brand-protector","provider":"akamai","name":"Brand Protector","category":"threat-intel","tagline":"Finds fake sites that copy your brand to trick your customers.","description":"Brand Protector looks for phishing sites and lookalike web addresses that pretend to be your company. If customers start getting emails that send them to a page that looks like yours but steals their passwords, this helps you find those sites so you can get them taken down. It is a good fit once your brand is known enough to be worth copying. Akamai's product index also lists related items called Brand Guardian and AI Brand Presence. Pricing is through Akamai sales.","useCases":["Find lookalike phishing domains","Take down fake brand sites","Monitor impersonation of your brand"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/brand-protector","source":"https://www.akamai.com/products"},{"id":"akamai-guardicore","provider":"akamai","name":"Akamai Guardicore Segmentation","category":"zero-trust","also":["networking"],"tagline":"Splits your servers into zones so one break-in can't spread everywhere.","description":"Akamai Guardicore Segmentation first maps how your applications talk to each other, then lets you set rules about which ones are allowed to connect. This is called microsegmentation. If an attacker or ransomware gets into one machine, those rules keep it from moving on to the rest, which limits the damage. It works across data centers and cloud workloads, and it can also set up separate zones for compliance. Pick it when you run many servers and want to contain problems. Pricing is through Akamai sales.","useCases":["Contain ransomware spreading between servers","Set up zones for compliance","Map and segment app traffic"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/akamai-guardicore-segmentation","source":"https://www.akamai.com/products"},{"id":"akamai-enterprise-application-access","provider":"akamai","name":"Enterprise Application Access","category":"zero-trust","tagline":"Let people reach private company apps without using a VPN.","description":"Enterprise Application Access gives your team and outside helpers a way into specific internal apps without a VPN, the older tool that connects people to your whole network. It checks who the person is first, then opens only the app they need. If you bring in contractors who need one admin tool, you can grant just that and nothing more. Pick it over a VPN when you want tighter, app-by-app access. This approach is known as ZTNA, short for zero trust network access. Pricing is through Akamai sales.","useCases":["Replace VPN access to internal apps","Give contractors access to specific apps","Limit staff to approved apps"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/enterprise-application-access","source":"https://www.akamai.com/products"},{"id":"akamai-secure-internet-access","provider":"akamai","name":"Secure Internet Access","category":"zero-trust","tagline":"Filters the websites your team visits to block malware and phishing.","description":"Secure Internet Access is a cloud service that checks where your team's internet traffic is going and blocks dangerous places. It filters by DNS (the internet's address lookup) and by web address, so staff are kept away from malware and phishing sites. If your people work from home or from cafes, you can apply the same web rules to everyone without installing hardware. It is built on Akamai's own recursive DNS, the service that looks up addresses for users. Pricing is through Akamai sales.","useCases":["Filter websites for employees","Block malware and phishing sites","Enforce web usage policy"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/secure-internet-access-enterprise","source":"https://www.akamai.com/products"},{"id":"akamai-workforce-protector","provider":"akamai","name":"Workforce Protector","category":"zero-trust","tagline":"Protects what your employees do in their web browser, including AI tools.","description":"Workforce Protector secures the browser your staff use all day, including the SaaS apps (online tools you subscribe to) and AI tools they open. If employees paste company data into an unapproved chatbot, a problem known as shadow AI, this helps you see it and control it. It can also help stop data from leaking out through the browser. Akamai's index lists it as formerly LayerX, which points to an acquisition. Pricing is through Akamai sales.","useCases":["Control staff use of AI tools","Prevent data loss in browsers","Secure SaaS use in the browser"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products","source":"https://www.akamai.com/products"},{"id":"akamai-mfa","provider":"akamai","name":"Akamai MFA","category":"identity","tagline":"Login checks that can't be tricked by fake sites, using your phone as a key.","description":"Akamai MFA adds a second check at login, on top of a password. It is built on FIDO2, an open standard for phishing-resistant sign-in, which means a fake login page can't easily trick people into handing over their access. Your employee's phone can act as the security key, so you don't have to buy and ship physical keys. Use it to protect your team's accounts, especially admins who can change important settings. Pick it over text-message codes when phishing is a real worry. Pricing is through Akamai sales.","useCases":["Protect logins with phishing-resistant MFA","Use a phone as a security key","Add stronger protection for admin accounts"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/mfa","source":"https://www.akamai.com/products"},{"id":"akamai-prolexic","provider":"akamai","name":"Prolexic","category":"ddos","tagline":"Soaks up huge floods of fake traffic before they reach your network.","description":"Prolexic defends against DDoS attacks, where attackers flood you with so much junk traffic that real visitors can't get through. It sends your traffic through Akamai's scrubbing centers, which filter out the bad and pass along the good. You can keep it always on, or turn it on only when an attack hits. If you run your own servers or data center and an outage would cost you money, this protects the network itself, not just a website. Pricing is through Akamai sales.","useCases":["Scrub floods aimed at data centers","Defend against large volume attacks","Turn protection on during attacks"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/prolexic-solutions","source":"https://www.akamai.com/products"},{"id":"akamai-ip-accelerator","provider":"akamai","name":"IP Accelerator","category":"networking","tagline":"Make games and connected devices feel faster and more reliable online.","description":"IP Accelerator speeds up internet traffic that isn't ordinary web pages, such as game data or messages from smart devices, by sending it across Akamai's global network. If you're building a multiplayer game, players far from your servers often feel lag, and this helps cut delays and dropped connections. It works with TCP and UDP, the two main ways apps send data over the internet, so it covers things a regular website cache can't help with. Pick it when your app talks in something other than web requests.","useCases":["Speed up TCP and UDP apps","Support gaming traffic latency","Improve IoT connection reliability"],"pricing":"Contact sales.","status":"ga","url":"https://www.akamai.com/products/ip-accelerator","source":"https://www.akamai.com/products"},{"id":"fastly-cdn","provider":"fastly","name":"Fastly CDN","category":"cdn","tagline":"Serve your site and APIs faster from locations close to your visitors.","description":"Fastly CDN is a worldwide network of servers that keeps copies of your content close to your visitors, so pages and API responses load faster. If your site shows things that change often, like prices or a news feed, you can clear old copies everywhere almost instantly. It's built on Varnish, a well-known caching tool, and you can tune its behavior with VCL, a small configuration language. Pick it over a basic CDN when you want fine control and live logs showing what's happening as it happens.","useCases":["Speed up dynamic websites","Cache responses from your API","Clear cached content within seconds"],"pricing":"Usage-based with a free trial. Contact sales for commits.","status":"ga","url":"https://www.fastly.com/products/cdn","source":"https://www.fastly.com/products"},{"id":"fastly-compute","provider":"fastly","name":"Fastly Compute","category":"edge-compute","also":["serverless"],"tagline":"Run your own code close to your users, written in Rust, JavaScript or Go.","description":"Fastly Compute lets you run small programs on servers spread around the world, near your visitors, instead of in one central place. If you want to personalize a page for each visitor or build a fast API without waiting on a distant server, this is where that code would live. Your code runs as WebAssembly, a compact format that is fast and safe, and every request gets its own isolated space that starts in microseconds. Pick it over a traditional server setup when quick response times matter most.","useCases":["Run custom code close to users","Build APIs that run without servers","Tailor pages for each visitor"],"pricing":"Usage-based. Free tier on trial.","status":"ga","url":"https://www.fastly.com/products/edge-compute","source":"https://www.fastly.com/products"},{"id":"fastly-ngwaf","provider":"fastly","name":"Next-Gen WAF","category":"waf","tagline":"A firewall for your site and APIs that blocks attacks, not real users.","description":"Next-Gen WAF, formerly called Signal Sciences, is a firewall for web apps (WAF stands for web application firewall) that checks incoming requests and blocks the harmful ones. If your app has a login form or a public API, it can stop attackers from slipping in malicious commands, a trick known as injection, or from abusing your API. It's known for low false positives, meaning it rarely blocks real customers by mistake. You can run it at the edge, on your own host, or in the cloud, so it fits different setups.","useCases":["Block injection attacks on your site","Stop abuse of your APIs","Protect web apps with fewer false alarms"],"pricing":"Contact sales. Tiered plans.","status":"ga","url":"https://www.fastly.com/products/web-application-api-protection","source":"https://www.fastly.com/products"},{"id":"fastly-bot-management","provider":"fastly","name":"Bot Management","category":"bots","tagline":"Stop fake traffic from bots that scrape, steal accounts or commit fraud.","description":"Bot Management spots automated programs visiting your site and stops the harmful ones before they reach your app. If you run a store or a login page, bots can try thousands of stolen passwords in a row, an attack called credential stuffing, or copy your prices and content by scraping. Detection happens at the edge, meaning on Fastly's servers close to the visitor, so bad traffic is dealt with early. Pick it when automated abuse is costing you money, customer trust or server capacity.","useCases":["Stop fraud from automated attacks","Block bots that stuff stolen logins","Stop bots from scraping content"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/bot-management","source":"https://www.fastly.com/products"},{"id":"fastly-ai-bot-management","provider":"fastly","name":"AI Bot Management","category":"bots","also":["ai-gateway"],"tagline":"See which AI crawlers visit your site, then allow, limit or block them.","description":"AI Bot Management identifies AI programs that read your website, such as crawlers collecting content for large language models (LLMs, the tech behind chatbots), and lets you decide what happens next. If you publish articles, courses or other original work, you can block the scrapers, let trusted ones in, or slow them down with a rate limit, which caps how many requests one visitor can make. Pick it over general bot tools when AI crawlers specifically are your worry and you want to control or meter their access.","useCases":["Spot AI scrapers and crawlers","Allow, limit, or block AI crawlers","Protect your content from AI scraping"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/bot-management","source":"https://www.fastly.com/products"},{"id":"fastly-ddos","provider":"fastly","name":"DDoS Protection","category":"ddos","tagline":"Automatic protection when floods of traffic try to knock your app offline.","description":"DDoS Protection keeps your site or API online when attackers send huge floods of fake requests to overwhelm it. It's always on and included in the Fastly edge, so you don't have to scramble to switch it on mid-attack. It focuses on layer 7 attacks, which are floods that look like normal web requests to your app. If you run a public API, that's the kind of flood that can slow things down for real customers. Pick it when you already deliver your traffic through Fastly.","useCases":["Defend APIs against flood attacks","Keep sites up during traffic floods","Get automatic protection that stays on"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/ddos-protection","source":"https://www.fastly.com/products"},{"id":"fastly-api-security","provider":"fastly","name":"API Security","category":"waf","tagline":"Find every API your company runs and spot people abusing them.","description":"API Security discovers the APIs your app exposes, the doorways that let other software talk to yours, and flags patterns that look like abuse. If your team has built quickly with many tools, some APIs may be forgotten or undocumented. These are called shadow APIs, and finding them is a common use for this service. Once you can see what's out there, you can spot suspicious activity and act on it. Pick it when you want to know what you're exposing before an attacker finds it first.","useCases":["Find APIs you didn't know existed","Flag suspicious API abuse patterns","Spot abuse patterns across your APIs"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/api-security","source":"https://www.fastly.com/products"},{"id":"fastly-client-side-protection","provider":"fastly","name":"Client-Side Protection","category":"waf","tagline":"Track and control the scripts on your pages to stop card data theft.","description":"Client-Side Protection keeps an inventory of every script running on your web pages and lets you manage them. Many sites load outside scripts for analytics, chat or ads, and a tampered one can quietly steal what customers type in, an attack called skimming. If you take card payments, it helps with PCI script compliance, the payment industry's rules for protecting card data. Pick it when your checkout or sign-up pages rely on third-party code that you don't fully control and want to keep watch over.","useCases":["Track every script on your pages","Block scripts that skim customer data","Support PCI compliance for scripts"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/fastly-client-side-protection","source":"https://www.fastly.com/products"},{"id":"fastly-image-optimizer","provider":"fastly","name":"Image Optimizer","category":"images","tagline":"Resize and convert images on the fly with a small change to the image link.","description":"Image Optimizer changes your images as they're requested, so you keep one original and get many sizes and formats from it. You add settings to the image's web address, and it resizes or converts the picture, then saves each version for fast reuse. If your app shows photos on both phones and large screens, you can serve right-sized ones in modern formats like WebP and AVIF, which are smaller files that still look good. Pick it when you don't want to create and store every version yourself. It's a usage-based add-on.","useCases":["Resize images on the fly","Convert images to WebP or AVIF","Serve sized images for each screen"],"pricing":"Usage-based add-on.","status":"ga","url":"https://www.fastly.com/products/image-optimization","source":"https://www.fastly.com/products"},{"id":"fastly-streaming-video","provider":"fastly","name":"Streaming Video","category":"video","tagline":"Deliver recorded videos to viewers in the right format for their device.","description":"Streaming Video delivers pre-recorded videos, known as video on demand or VOD, to viewers around the world. It does the packaging at the edge, which means it prepares each video in the format a player needs on servers close to the viewer. If you run a course platform or a library of tutorials, you don't have to create and store every version yourself. Pick it when you have a catalog of recorded videos rather than live broadcasts. Pricing is based on how much you use it.","useCases":["Deliver on-demand video libraries","Convert video into playable formats","Stream movies and clips to viewers"],"pricing":"Usage-based.","status":"ga","url":"https://www.fastly.com/products/streaming-media","source":"https://www.fastly.com/products"},{"id":"fastly-live-streaming","provider":"fastly","name":"Live Streaming","category":"video","tagline":"Send live video to lots of viewers at once without it falling over.","description":"Live Streaming delivers live video to large audiences by spreading the load across Fastly's servers close to viewers. If you host a webinar, a product launch or a live event, thousands of people may join at the same moment, and this helps keep the stream smooth for all of them. It's tuned for HLS and DASH, two common ways of cutting live video into small pieces that play in almost any video player. Pick it when your video is happening right now, rather than sitting in a library. Pricing is based on usage.","useCases":["Stream live events to viewers","Deliver live video to large audiences","Broadcast live events at scale"],"pricing":"Usage-based.","status":"ga","url":"https://www.fastly.com/products/live-streaming","source":"https://www.fastly.com/products"},{"id":"fastly-media-shield","provider":"fastly","name":"Media Shield","category":"video","tagline":"Cuts repeat requests to your video source when you use several CDNs.","description":"Media Shield is a middle layer that sits in front of your video storage and bundles many requests for the same file into fewer ones. If you stream video through more than one CDN (a delivery network that serves files from locations near your viewers), each CDN would normally ask your storage for the same content. Media Shield consolidates those requests, so your origin, the server where your videos live, sees less traffic. Pick it when you split streaming across several CDNs and want to lighten the load on your own setup. Pricing is available through Fastly sales.","useCases":["Cut origin traffic for video","Run video across several CDNs","Lower load on your main servers"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/streaming-media/media-shield","source":"https://www.fastly.com/products"},{"id":"fastly-otfp","provider":"fastly","name":"On-the-Fly Packager","category":"video","tagline":"Prepares video for streaming the moment a viewer asks for it.","description":"On-the-Fly Packager converts your original video files into streaming formats at the moment someone presses play, instead of ahead of time. If your app has a library of videos on demand, like courses or recorded webinars, you can skip the offline packaging step, which is a separate job that prepares every format in advance. That means less pipeline to build and look after. Choose it when you want your video setup to stay simple and don't want to pre-process every file. Pricing is available through Fastly sales.","useCases":["Package video when requested","Skip preparing video files ahead of time","Serve video without pre-processing"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/perform/otfp","source":"https://www.fastly.com/products"},{"id":"fastly-load-balancer","provider":"fastly","name":"Fastly Load Balancer","category":"load-balancing","tagline":"Spreads traffic across your servers and reroutes if one goes down.","description":"Fastly Load Balancer splits incoming visitors across several servers and regularly checks that each one is working. If your app runs on two servers, or in two different clouds, and one stops responding, it can send people to the one that is still healthy. It runs at the edge, meaning on Fastly's servers close to your visitors, and gives you visibility into what is happening. It is a natural pick if you already use Fastly's CDN, since it is included with CDN plans.","useCases":["Spread traffic across several servers","Switch to backup servers automatically","Check server health continuously"],"pricing":"Included with CDN plans.","status":"ga","url":"https://www.fastly.com/products/load-balancing","source":"https://www.fastly.com/products"},{"id":"fastly-tls","provider":"fastly","name":"TLS Encryption","category":"certificates","tagline":"Gives your custom domain HTTPS, with certificates handled for you.","description":"TLS Encryption is what puts the padlock and the https in front of your website address. Fastly issues and manages the certificates (digital proof that your site is really yours) and does the encryption work at the edge, on servers close to your visitors. If you launch your product on your own domain, like yourbrand.com, this gets HTTPS running without you juggling certificates yourself. Pick it when you already use Fastly and want one less thing to maintain. Whether it is included or an add-on depends on your plan.","useCases":["Serve custom domains over HTTPS","Issue and renew certificates automatically","Offload encryption from your servers"],"pricing":"Included or add-on depending on plan.","status":"ga","url":"https://www.fastly.com/products/tls-encryption","source":"https://www.fastly.com/products"},{"id":"fastly-origin-connect","provider":"fastly","name":"Origin Connect","category":"networking","tagline":"A private line between your servers and Fastly.","description":"Origin Connect is a dedicated connection between your data center or cloud account and Fastly's edge, the servers close to your visitors. If your app's data sits in a cloud and you would rather not send that traffic over the open internet, this gives it a private route instead. It is different from a standard setup, where Fastly reaches your origin (the place your content comes from) over public connections. Choose it when privacy or a controlled path matters to you. Pricing is available through Fastly sales.","useCases":["Connect your servers privately to Fastly","Send server traffic over a private link","Keep server traffic off the public internet"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/ways-to-connect","source":"https://www.fastly.com/products"},{"id":"fastly-dedicated-ips","provider":"fastly","name":"Dedicated IP Addresses","category":"networking","tagline":"Your own fixed IP addresses on Fastly, or bring ones you already own.","description":"Dedicated IP Addresses give your domains their own reserved IP addresses, the numeric addresses that computers use to find your site, instead of sharing them. If a partner or customer only lets approved addresses connect to their systems, which is called allowlisting, fixed addresses make that possible. You can also bring your own IPs, known as BYOIP, if you already own a set. Pick this when someone needs a stable address from you. Pricing is available through Fastly sales.","useCases":["Give partners fixed addresses to allow","Use your own IP addresses","Reserve IPs for your domains"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/ip-addresses","source":"https://www.fastly.com/products"},{"id":"fastly-kv-store","provider":"fastly","name":"KV Store","category":"kv-cache","tagline":"Simple lookup storage that lives close to your visitors.","description":"KV Store is a key-value database, which works like a labeled list: you look up a name and get back its saved value. Your data is copied around the world, so reads happen on servers near your visitors. If you want to switch features on and off for certain users (often called feature flags) or keep a lookup table of redirects, it fits well. It works with Compute, Fastly's way of running your code at the edge, and with VCL, Fastly's configuration language. Pricing is based on usage.","useCases":["Store feature flags for quick reads","Keep lookup tables close to users","Keep data replicated worldwide for fast reads"],"pricing":"Usage-based.","status":"ga","url":"https://www.fastly.com/products/kv-store","source":"https://www.fastly.com/products"},{"id":"fastly-config-store","provider":"fastly","name":"Config Store","category":"kv-cache","tagline":"Change small settings at the edge without redeploying your code.","description":"Config Store holds small pieces of configuration that your Compute code reads while it runs. Compute is Fastly's way of running your code on servers close to your visitors. If you need to change a setting, like which backend to use or a message to show, you can update it here without publishing a new version of your code. Pick it over KV Store when what you have is a handful of settings rather than a growing pile of data. It is included with Compute.","useCases":["Change settings without redeploying code","Hold small runtime configuration values","Update config while apps keep running"],"pricing":"Included with Compute.","status":"ga","url":"https://www.fastly.com/documentation/guides/concepts/edge-state/dynamic-config/","source":"https://www.fastly.com/products"},{"id":"fastly-secret-store","provider":"fastly","name":"Secret Store","category":"secrets","tagline":"Keeps API keys and passwords safe for your Fastly edge code.","description":"Secret Store is an encrypted place to keep sensitive values like API keys and tokens, so they stay out of your code. If your Compute code, which runs on Fastly's servers close to your visitors, needs a password to fetch data from your private backend, it can read that credential from here when it runs. That is safer than pasting the secret into your code, where it could leak into a repository or a screenshot. Pick it whenever your edge code needs to prove who it is. It is included with Compute.","useCases":["Keep API keys encrypted and safe","Store tokens that code can read","Hold origin login credentials securely"],"pricing":"Included with Compute.","status":"ga","url":"https://www.fastly.com/documentation/guides/concepts/edge-state/dynamic-config/","source":"https://www.fastly.com/products"},{"id":"fastly-object-storage","provider":"fastly","name":"Fastly Object Storage","category":"object-storage","tagline":"Store images and files, with tools that work like Amazon S3.","description":"Fastly Object Storage is a place to keep files such as images, videos and downloads. It is S3-compatible, meaning tools built for Amazon's S3 file storage can work with it. If your app has a website with lots of images or downloadable files, you can store them here and let Fastly's CDN, a network that delivers files from servers near your visitors, serve them. It is designed to pair with the Fastly CDN and Compute, so it makes the most sense if you already use Fastly. Pricing is based on usage.","useCases":["Store files in S3-compatible storage","Hold static assets behind the CDN","Use as origin for static files"],"pricing":"Usage-based.","status":"ga","url":"https://www.fastly.com/documentation/guides/platform/object-storage/","source":"https://www.fastly.com/products"},{"id":"fastly-ai-accelerator","provider":"fastly","name":"Fastly AI Accelerator","category":"ai-gateway","tagline":"Saves money on AI chatbot calls by reusing answers to similar questions.","description":"Fastly AI Accelerator remembers answers from the large language models (LLMs, the AI behind chatbots) your app calls. When a new question means about the same as an earlier one, it returns the saved answer instead of asking the model again. If your support bot gets the same question in a hundred different wordings, this can cut both your bill and the wait time. It works with apps that already use OpenAI-style APIs. Pricing is based on usage, so contact Fastly sales.","useCases":["Cache similar AI prompts to save","Lower costs of repeated LLM calls","Speed up responses from AI APIs"],"pricing":"Usage-based. Contact sales.","status":"ga","url":"https://www.fastly.com/products/ai-accelerator","source":"https://www.fastly.com/products"},{"id":"fastly-fanout","provider":"fastly","name":"Fanout","category":"realtime","tagline":"Push live updates to your users' screens the moment something happens.","description":"Fanout lets your app push messages to many users at once, so their screens update without a refresh. It supports WebSockets, SSE and long polling, which are three common ways to keep a live connection open between a browser and a server. If you are building a chat, a live score board or a notification feed, this handles the heavy lifting of holding all those connections. It runs through Fastly's edge using the GRIP protocol, a standard way for the edge to hold connections for your server. Pricing is based on usage.","useCases":["Push live updates to browsers","Power chat messages in real time","Send messages over WebSockets globally"],"pricing":"Usage-based.","status":"ga","url":"https://www.fastly.com/products/fanout","source":"https://www.fastly.com/products"},{"id":"fastly-logging","provider":"fastly","name":"Fastly Logging","category":"observability","tagline":"Stream a live record of requests to your storage or monitoring tools.","description":"Fastly Logging sends a record of every request that reaches your site to the place you keep your data, within seconds. If you need to investigate suspicious traffic after a security scare, or want visitor activity flowing into an analytics dashboard, you can stream the logs straight there instead of collecting them by hand. It supports more than 30 destinations, so it probably works with tools you already use. It is part of Fastly, a service that delivers your site from servers around the world, and pricing is listed as included.","useCases":["Stream request logs within seconds","Send logs to your analytics tools","Investigate security incidents with logs"],"pricing":"Included.","status":"ga","url":"https://www.fastly.com/products/logging","source":"https://www.fastly.com/products"},{"id":"fastly-metrics","provider":"fastly","name":"Metrics and Inspectors","category":"observability","tagline":"Live traffic stats, with a closer look at each domain and your own servers.","description":"Metrics and Inspectors shows how much traffic your site gets and how it is performing, broken down by domain and by origin (the server where your original site or app lives). If your site suddenly feels slow or errors spike, you can check whether the trouble is in the incoming traffic or on your own server. It gives you real-time stats plus two views, Domain Inspector and Origin Inspector, so you can dig in without extra tools. Pricing is listed as included.","useCases":["Monitor traffic for each domain","Check whether origin servers are healthy","Debug slow or failing traffic"],"pricing":"Included.","status":"ga","url":"https://www.fastly.com/products/metrics","source":"https://www.fastly.com/products"},{"id":"fastly-enterprise-serverless","provider":"fastly","name":"Enterprise Serverless","category":"serverless","also":["edge-compute"],"tagline":"Run code near your users, with written support guarantees for big companies.","description":"Enterprise Serverless is Fastly's package for running your code without managing servers yourself. It bundles Compute with edge state (data stored close to your visitors) and is built on open standards, meaning shared rules rather than a locked-in setup. If your code currently runs in one central cloud location and you want to move it closer to users, this is the use case it is built for. Pick it over the basic products when you want SLAs, which are written service promises. Pricing is through sales.","useCases":["Run server code at the edge","Store data close to users","Get service guarantees for enterprise contracts"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/products/edge-compute","source":"https://www.fastly.com/products"},{"id":"fastly-managed-security","provider":"fastly","name":"Managed Security","category":"waf","tagline":"Experts run your site's firewall and attack protection 24/7.","description":"Managed Security means Fastly's own security operations team runs your protection for you, around the clock. That covers a WAF (a web application firewall, which blocks harmful requests), bot protection, and DDoS defense (stopping floods of traffic meant to knock your site offline). If you do not have anyone on staff to tune these settings, you can hand that job over instead of learning it. Choose it over doing it yourself when you want people watching your site at all hours. Pricing is through sales.","useCases":["Have experts run firewall and bot protection","Stop attacks around the clock","Skip in-house security tuning work"],"pricing":"Contact sales.","status":"ga","url":"https://www.fastly.com/services/managed-security-service","source":"https://www.fastly.com/products"},{"id":"zscaler-zia","provider":"zscaler","name":"Zscaler Internet Access (ZIA)","category":"zero-trust","tagline":"A cloud checkpoint that screens your team's web and SaaS traffic.","description":"Zscaler Internet Access sits between your team and the internet, checking web and SaaS (online software) traffic in the cloud as it passes through. It includes URL filtering, threat prevention, DLP (tools that stop sensitive data leaving), and sandboxing (testing suspicious files in a safe space). If your company still sends staff traffic through proxy boxes in an office, this can replace them. It also lets you control which SaaS apps people use. Pricing comes in tiered bundles, and you contact sales for a quote.","useCases":["Filter websites and block risky downloads","Inspect web and SaaS traffic inline","Replace on-prem proxy servers"],"pricing":"Contact sales. Tiered bundles.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zscaler-internet-access","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zpa","provider":"zscaler","name":"Zscaler Private Access (ZPA)","category":"zero-trust","tagline":"Let your team reach private apps safely, without a VPN.","description":"Zscaler Private Access connects your people to your company's private apps without a traditional VPN (a tunnel into your whole network). It acts as a go-between, so the apps are never exposed to the network itself. This is called zero trust, meaning nobody gets in just for being on the network. If your team is tired of clunky VPN logins, it can replace them. It also helps after a merger or acquisition, when people need access to apps from another company. Pricing is through sales.","useCases":["Give staff access to private apps","Skip the VPN for remote access","Connect systems after a company merger"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zscaler-private-access","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zdx","provider":"zscaler","name":"Zscaler Digital Experience (ZDX)","category":"observability","tagline":"Find out why an app feels slow: your network, your device, or the app.","description":"Zscaler Digital Experience watches what your team actually experiences when they use apps, from their device all the way to the service. It does this with endpoint probes (small checks running on each person's device) plus path data from Zscaler. If someone tells your help desk that everything is slow, you can see whether the cause is their Wi-Fi, their laptop, or the app, and skip the guessing. It is a good fit for sorting support tickets and chasing app slowness. Pricing is through sales.","useCases":["Find where app slowness starts","Triage help desk tickets faster","Monitor user experience end to end"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zscaler-digital-experience","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-data-protection","provider":"zscaler","name":"Data Protection","category":"zero-trust","tagline":"Stop sensitive data leaking out through web, SaaS, email and devices.","description":"Data Protection helps keep private information from leaving your company by accident or on purpose. It covers web traffic, SaaS apps (online software), email, and employee devices. It includes DLP (data loss prevention) and CASB (a tool that watches how your team uses cloud apps), both inline, which checks data as it moves, and out-of-band, which scans what is already stored. It also checks SaaS settings for mistakes, such as a folder shared with the whole internet. Pricing is through sales.","useCases":["Block sensitive data from leaving","Find risky SaaS settings","Stop data leaks through email"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/data-protection","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zero-trust-branch","provider":"zscaler","name":"Zero Trust Branch","category":"networking","tagline":"Connect offices, campuses and factories securely without a site firewall.","description":"Zero Trust Branch links your physical locations, such as offices, campuses and factories, to Zscaler without needing a firewall box at each site. It is meant to replace SD-WAN, a software-managed network that ties locations together. If you run several offices, it gives each one secure connectivity without extra hardware to maintain. It also covers factory OT (the equipment that runs production lines), which is harder to protect than regular computers. Pricing is through sales.","useCases":["Connect branch offices securely","Secure factory floor connections","Replace SD-WAN gear at sites"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zero-trust-branch","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zero-trust-cloud","provider":"zscaler","name":"Zero Trust Cloud","category":"zero-trust","tagline":"Secure the traffic your cloud workloads send to the internet and each other.","description":"Zero Trust Cloud protects the communication of your workloads, meaning the apps and servers you run in the cloud. It secures traffic going out to the internet and traffic moving between different clouds. If your servers in one cloud need to call outside services or talk to another cloud, this puts a security check on those connections. It is most useful when you spread across more than one cloud provider and want one way to control outbound traffic. Pricing is through sales.","useCases":["Secure traffic from cloud workloads","Control outbound internet traffic from workloads","Secure traffic between clouds"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zero-trust-cloud","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zero-trust-browser","provider":"zscaler","name":"Zero Trust Browser","category":"zero-trust","tagline":"Safe browsing and app access on any device, even ones you don't manage.","description":"Zero Trust Browser uses browser isolation, which means web pages and apps load on a remote server and only a safe view reaches the person's device. That keeps risky content away from laptops and phones you do not control. If you work with contractors, or let people use their own devices (often called BYOD), you can give them access to private apps without installing anything heavy or trusting their hardware. It also covers secure everyday web browsing. Pricing is through sales.","useCases":["Give contractors safe browser access","Let unmanaged devices browse safely","Isolate browsing on personal devices"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zero-trust-browser","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-cloud-sandbox","provider":"zscaler","name":"Cloud Sandbox","category":"threat-intel","tagline":"Test unknown files in a safe bubble before they reach your team.","description":"Cloud Sandbox opens files it has never seen before in an isolated space and watches what they do before letting them through. This helps stop zero-day malware, which is brand-new harmful software that security tools do not recognize yet. It works inline, meaning files are checked as they travel to your team, and it uses AI verdicts to decide whether each one is safe. If you already use Zscaler Internet Access, it is included in those tiers.","useCases":["Test unknown files before opening","Catch zero-day malware in files","Score files with AI verdicts"],"pricing":"Included in ZIA tiers.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/cloud-sandbox","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zero-trust-firewall","provider":"zscaler","name":"Zero Trust Firewall","category":"zero-trust","also":["networking"],"tagline":"A cloud firewall that checks non-web traffic on every port and protocol.","description":"Zero Trust Firewall is a firewall that runs in the cloud instead of in a box at your office, and it applies security rules to traffic that is not ordinary web browsing. If your company has several branch offices, each with its own firewall device, you could use this to retire those devices and manage the rules in one place. It covers all ports and protocols (the different channels and languages apps use to talk over a network) and includes IPS, a feature that spots and blocks known attack patterns. Pricing is available by contacting Zscaler sales.","useCases":["Apply firewall rules to all ports","Filter non-web traffic with IPS policy","Retire branch firewall hardware"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zero-trust-firewall","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-zero-trust-gateway","provider":"zscaler","name":"Zero Trust Gateway","category":"networking","tagline":"One checkpoint that secures traffic from your servers and devices.","description":"Zero Trust Gateway is a single checkpoint that network traffic from your workloads passes through. A workload is simply a program running on a server or in the cloud. If you want to control which outside services your app's servers are allowed to reach, you would use this for egress control, which means rules for outbound traffic. It is built to secure all workload communications through one gateway, so you can pick it when you want one place to set and check those outbound rules. Pricing is available by contacting Zscaler sales.","useCases":["Route workload traffic through one gateway","Control device and workload traffic","Secure all workload communications"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zero-trust-gateway","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-business-continuity","provider":"zscaler","name":"Business Continuity","category":"zero-trust","tagline":"Keeps security and access working if Zscaler or your internet provider fails.","description":"Business Continuity is a safety net for your Zscaler setup. It provides failover, meaning an automatic switch to a backup, so security rules keep being enforced for ZIA (Zscaler's service for safe internet browsing) and ZPA (its service for reaching private company apps). If your team depends on those tools all day, this is what keeps people working during a Zscaler or ISP (internet provider) outage. Pick it when downtime would be costly and you want resilience built in. Pricing is available by contacting Zscaler sales.","useCases":["Keep access working during outages","Maintain policy during ISP failures","Fail over security during Zscaler outages"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/business-continuity","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-privileged-remote-access","provider":"zscaler","name":"Privileged Remote Access","category":"zero-trust","tagline":"Let admins and vendors in through a browser, with sessions recorded.","description":"Privileged Remote Access lets people with special access, like system admins or outside vendors, log in to your systems through a web browser. It is clientless, which means nobody has to install extra software first. If a contractor needs to fix something on your equipment, you can give them a browser session and keep a recording of what they did. It also suits OT admin, meaning managing operational technology like factory or industrial machines. Pricing is available by contacting Zscaler sales.","useCases":["Give admins browser-based privileged sessions","Let vendors connect without client software","Record privileged admin sessions"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/privileged-remote-access","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-cellular","provider":"zscaler","name":"Zscaler Cellular","category":"zero-trust","also":["iot"],"tagline":"Security rules for cellular devices and IoT gadgets.","description":"Zscaler Cellular brings zero trust policy to devices that connect over cellular networks. Zero trust means no connection is trusted by default, and every one is checked against your rules. It covers cellular traffic and SIM-based devices, the ones with a SIM card like a phone has. If you run an IoT fleet, such as connected sensors, trackers or vending machines, you can use this to apply the same security rules to them as to your laptops. Pricing is available by contacting Zscaler sales.","useCases":["Apply zero trust to cellular devices","Secure SIM-based IoT devices","Protect IoT device fleets"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions/zscaler-cellular","source":"https://www.zscaler.com/products-and-solutions"},{"id":"zscaler-endpoint-context","provider":"zscaler","name":"Endpoint Context","category":"zero-trust","tagline":"Lets security rules consider which program on a device sent the traffic.","description":"Endpoint Context adds extra detail to the security rules Zscaler applies to traffic as it passes through. Normally a rule might only see where traffic is going. With this, a rule can also consider which process, meaning which running program on a laptop or other device, created that traffic. If you want to allow one app to reach something while blocking another app on the same machine, this is how you would do it. It is for app-aware policy. Pricing is available by contacting Zscaler sales.","useCases":["Set policy by the process sending traffic","Make policy aware of apps","Tell which process generated traffic"],"pricing":"Contact sales.","status":"ga","url":"https://www.zscaler.com/products-and-solutions","source":"https://www.zscaler.com/products-and-solutions"},{"id":"vercel-deployments","provider":"vercel","name":"Vercel Deployments","category":"hosting","tagline":"Push your code and get a live site, plus a preview link for every change.","description":"Vercel Deployments is a hosting and build service that turns your code into a live website. Every time you push a change using Git (the tool that tracks your code), it creates an immutable deployment, a saved version that never changes, and gives it a preview link. If you are building with Next.js, a popular framework for websites, this is the natural home since Vercel is where it comes from. There is a free Hobby plan, and Pro starts at $20 per user per month.","useCases":["Deploy each git push automatically","Get preview links for every change","Host Next.js and frontend apps"],"pricing":"Free Hobby plan. Pro from $20/user/month.","status":"ga","url":"https://vercel.com/docs/deployments","source":"https://vercel.com/docs/products"},{"id":"vercel-functions","provider":"vercel","name":"Vercel Functions (Fluid compute)","category":"serverless","tagline":"Run backend code that bills for the time it is actually working.","description":"Vercel Functions run your backend code without you managing a server. They use Fluid compute, which reuses one running instance for several requests and bills for active CPU time, not time spent waiting. If your app streams answers from an AI model, much of each request is spent waiting, so this can suit you well. It also handles API routes and SSR, which means building pages on the server before sending them. Pricing is usage-based, covering active CPU, memory and invocations.","useCases":["Run API routes without servers","Render pages on the server","Stream AI responses efficiently"],"pricing":"Usage-based. Active CPU, memory and invocations.","status":"ga","url":"https://vercel.com/docs/functions","source":"https://vercel.com/docs/products"},{"id":"vercel-cdn","provider":"vercel","name":"Vercel CDN","category":"cdn","tagline":"Serves your site's files from places near your visitors, with caching.","description":"Vercel CDN is a global delivery network that keeps copies of your site's files and saved responses close to your visitors, so pages load faster. If your site has images, scripts and pages that rarely change, this is what serves them quickly around the world. It also supports ISR caching, which refreshes saved pages without a full rebuild. The billing metric once called Edge Requests is now called CDN Requests. Pricing is usage-based, and Pro plans have a Flat Rate CDN option.","useCases":["Deliver static files from nearby","Cache responses at the edge","Speed up static asset delivery"],"pricing":"Usage-based with Flat Rate CDN option on Pro.","status":"ga","url":"https://vercel.com/docs/cdn","source":"https://vercel.com/docs/products"},{"id":"vercel-routing-middleware","provider":"vercel","name":"Routing Middleware","category":"edge-compute","tagline":"Code that runs before a request reaches your page, for redirects and logins.","description":"Routing Middleware is a small piece of code that runs at the edge, meaning on servers near your visitor, before a request is sent to the right page. If you want to send visitors to a local version of your site based on their country, or block people who are not logged in, this is the place to do it. It handles rewrites, redirects and auth checks, which are login and permission checks. It replaces the older name, Edge Middleware. Pricing is usage-based.","useCases":["Check logins before routing requests","Redirect visitors based on location","Rewrite URLs before pages load"],"pricing":"Usage-based.","status":"ga","url":"https://vercel.com/docs/routing-middleware","source":"https://vercel.com/docs/products"},{"id":"vercel-ai-sdk","provider":"vercel","name":"AI SDK","category":"devtools","also":["agents"],"tagline":"Free toolkit for building chatbots and AI agents in TypeScript.","description":"AI SDK is a free, open source toolkit for building AI features into apps written in TypeScript, a typed flavor of JavaScript. If you are making a chatbot that types out answers as they arrive, it handles streaming text, tool calls, structured output and agents, which are AI helpers that take actions. It works across model providers, so you are not tied to one company's AI. It also includes support for coding agents. There is no cost to use it.","useCases":["Build streaming chatbot interfaces","Add tool calls to AI apps","Create agents across model providers"],"pricing":"Free, open source.","status":"ga","url":"https://ai-sdk.dev/","source":"https://vercel.com/docs/products"},{"id":"vercel-ai-gateway","provider":"vercel","name":"Vercel AI Gateway","category":"ai-gateway","also":["ai-models"],"tagline":"One API key for 400+ models, with fallbacks and spend tracking.","description":"Vercel AI Gateway gives you one endpoint and one API key for more than 400 AI models from many providers. If your app uses several models, you can switch between them without juggling accounts and bills. It offers fallbacks (a backup model if one fails), budgets, per-user limits, logs and zero data retention. It passes token prices through with no markup, so you pay each model's list price. That makes it a good fit when you want cost controls.","useCases":["Use one key for many models","Set budgets and per-user limits","Fall back when a model fails"],"pricing":"Pay model list prices. No markup.","status":"ga","url":"https://vercel.com/ai-gateway","source":"https://vercel.com/docs/products"},{"id":"vercel-v0","provider":"vercel","name":"v0","category":"ai-apps","also":["devtools"],"tagline":"Describe an app in plain words and get a deployable first version.","description":"v0 is an AI tool from Vercel that turns a written description into screens and full apps. If you want to test an idea before hiring a developer, you can describe a dashboard or a landing page and get a first version to react to. It outputs Next.js, a popular framework for building websites, so the result can go live on Vercel with one click. Pick it over a blank code editor when you want a fast prototype or an internal tool. There is a free tier, and paid plans start at $20 a month.","useCases":["Turn prompts into UI mockups","Build full-stack apps from prompts","Deploy internal tools quickly"],"pricing":"Free tier. Paid plans from $20/month.","status":"ga","url":"https://v0.app/","source":"https://vercel.com/docs/products"},{"id":"vercel-sandbox","provider":"vercel","name":"Vercel Sandbox","category":"agents","tagline":"A locked-off computer for safely running code your AI writes.","description":"Vercel Sandbox gives you a short-lived, isolated Linux computer where code can run without touching the rest of your system. If your app has an AI agent that writes code, or you let users run their own scripts, this is where you run it safely. Each sandbox is a microVM, a tiny virtual machine, and it supports snapshots, 64 GB of storage and every Vercel region. Choose it over running that code on your own server when you can't trust what the code might do. Many coding agents already use it, and you pay for usage.","useCases":["Run AI-generated code safely","Run untrusted code in isolation","Test coding agents with snapshots"],"pricing":"Usage-based. Active CPU, memory, creation.","status":"ga","url":"https://vercel.com/sandbox","source":"https://vercel.com/docs/products"},{"id":"vercel-queues","provider":"vercel","name":"Vercel Queues","category":"queues","tagline":"A waiting line for your app's tasks, so slow work runs in the background.","description":"Vercel Queues is a message queue, a waiting line where your app drops tasks to be handled later. If someone signs up and you need to send emails and update other tools, you can put each job in the queue instead of making the person wait. It can also split one big job into many smaller ones that run side by side. It suits you if your app already lives on Vercel. It is in beta and began as a limited beta, so check the docs for current availability. Pricing is usage-based.","useCases":["Hold background jobs until processed","Spread work across many workers","Queue messages for later processing"],"pricing":"Usage-based.","status":"beta","url":"https://vercel.com/docs/queues","source":"https://vercel.com/docs/products"},{"id":"vercel-workflow","provider":"vercel","name":"Vercel Workflow","category":"workflows","tagline":"Write multi-step jobs as code that picks up where it left off.","description":"Vercel Workflow lets you write long, multi-step jobs as ordinary TypeScript functions, a typed flavor of JavaScript. If you run an onboarding flow that sends a welcome email, waits two days, then follows up, or an AI agent that loops for a long time, the job keeps going through deploys and failures. It comes as an open source SDK, a code toolkit, plus a managed backend that Vercel runs for you. Pick it over a simple background job when the task has many steps and must not lose its place. Pricing is usage-based.","useCases":["Run agent loops that survive deploys","Run multi-step jobs that resume after failures","Guide new users through onboarding flows"],"pricing":"Usage-based.","status":"ga","url":"https://vercel.com/docs/workflow","source":"https://vercel.com/docs/products"},{"id":"vercel-blob","provider":"vercel","name":"Vercel Blob","category":"object-storage","tagline":"Store the files your users upload, like photos and videos.","description":"Vercel Blob is file storage for your app. If your app lets people upload profile photos, or you serve images and videos, Blob holds those files and delivers them quickly. It is a natural pick if you already host on Vercel, since your files live alongside your app. Since September 2026, every plan can have unlimited stores. You pay for what you use, for both storage and transfer.","useCases":["Store user uploads in one place","Host media files for your app","Serve fast file downloads to visitors"],"pricing":"Usage-based storage and transfer.","status":"ga","url":"https://vercel.com/docs/vercel-blob","source":"https://vercel.com/docs/products"},{"id":"vercel-edge-config","provider":"vercel","name":"Edge Config (Global Config)","category":"kv-cache","tagline":"Tiny settings your app reads instantly, like feature on/off switches.","description":"Edge Config is a small data store, copied to servers around the world, so reading from it is very fast. If you want to flip a new feature on or off, or keep a list of page redirects, you can keep those settings here. It is built for data that is read constantly and changed rarely, so it is not the place for user records. Vercel's docs now label this area Global Config. It is included with allowances, then priced by usage above them.","useCases":["Serve feature toggles with low delay","Keep redirect lists for your site","Store settings that are read often"],"pricing":"Included with allowances, usage-based above.","status":"ga","url":"https://vercel.com/docs/edge-config","source":"https://vercel.com/docs/products"},{"id":"vercel-marketplace-storage","provider":"vercel","name":"Marketplace Storage","category":"sql","also":["kv-cache"],"tagline":"Add a database to your app in one click, billed through Vercel.","description":"Marketplace Storage lets you add databases and other data stores from partner companies without leaving Vercel. If your app needs somewhere to keep user accounts, you can add Postgres, a popular database, or Redis, a fast store for temporary data, from vendors such as Neon, Upstash and Supabase. Amazon S3, a widely used file storage service, was added in October 2026. Pick it when you want quick setup and a single bill. Pricing follows each provider, billed through Vercel.","useCases":["Add a Postgres or Redis database","Connect a database from Neon or Supabase","Pay for databases through Vercel billing"],"pricing":"Provider pricing, billed via Vercel.","status":"ga","url":"https://vercel.com/marketplace","source":"https://vercel.com/docs/products"},{"id":"vercel-firewall","provider":"vercel","name":"Vercel Firewall","category":"waf","also":["ddos"],"tagline":"Block attacks and abusive traffic before they reach your app.","description":"Vercel Firewall is a shield in front of your site. DDoS mitigation, which absorbs floods of fake traffic meant to knock you offline, is always on. On top of that you can set WAF rules, filters that decide which requests are allowed, plus IP blocking and rate limiting. If someone is hammering your API, a rate limit caps how often they can call it. It is included with Vercel, though advanced features need the Pro or Enterprise plan.","useCases":["Block abusive traffic before it reaches your app","Rate limit requests to your API","Block attacks with custom firewall rules"],"pricing":"Included. Advanced features on Pro and Enterprise.","status":"ga","url":"https://vercel.com/docs/vercel-firewall","source":"https://vercel.com/docs/products"},{"id":"vercel-botid","provider":"vercel","name":"BotID","category":"bots","tagline":"Spot bots on your key pages without making real people solve puzzles.","description":"BotID works out whether a visitor is a real person or a bot, without showing a CAPTCHA, those click-the-traffic-lights puzzles. If fake accounts keep signing up or bots keep hitting your checkout, you can protect those specific pages and leave the rest alone. A Deep Analysis option is available when you want a closer look. Choose it over a general firewall rule when the problem is bots, not just heavy traffic. You pay per verified request.","useCases":["Protect checkout from bots","Stop bots from abusing signup","Detect bots without showing challenges"],"pricing":"Usage-based per verified request.","status":"ga","url":"https://vercel.com/docs/botid","source":"https://vercel.com/docs/products"},{"id":"vercel-bot-management","provider":"vercel","name":"Bot Management","category":"bots","tagline":"Ready-made rules to block scrapers and control AI crawlers.","description":"Bot Management is a set of ready-made bot rules inside Vercel Firewall. If a scraper keeps copying your content, or you want a say in which AI crawlers, programs that read websites to collect data, can visit, you can switch on rules instead of writing your own. It is the quicker pick when you want general bot cleanup across your whole site. For guarding one key action such as signup, BotID is the closer fit. It is included with Firewall tiers.","useCases":["Block scrapers with managed rule sets","Control which AI crawlers can visit","Apply bot rules inside your firewall"],"pricing":"Included with Firewall tiers.","status":"ga","url":"https://vercel.com/docs/bot-management","source":"https://vercel.com/docs/products"},{"id":"vercel-deployment-protection","provider":"vercel","name":"Deployment Protection","category":"identity","tagline":"Put a password or company login in front of your preview links.","description":"Deployment Protection controls who can open your app's web addresses. If you share a work-in-progress version with a client, you can require a password or SSO, single sign-on with a work account, so only invited people see it. It covers both preview and production versions of your app. Since September 2026, production protection is free on every plan. It is included with Vercel, though some modes require the Pro plan.","useCases":["Keep preview sites private to your team","Gate previews with a password or SSO","Limit who can open deployment URLs"],"pricing":"Included. Some modes require Pro.","status":"ga","url":"https://vercel.com/docs/deployment-protection","source":"https://vercel.com/docs/products"},{"id":"vercel-web-analytics","provider":"vercel","name":"Vercel Web Analytics","category":"analytics","tagline":"See how many people visit your site, without cookies.","description":"Vercel Web Analytics shows you how many people visit your site and which pages they view. If you just launched a landing page and want to know whether your campaign is bringing in traffic, this gives you the numbers. It is cookieless, meaning it doesn't place tracking files in visitors' browsers, which keeps it privacy-friendly. It connects to your deployments, so it suits you if you host on Vercel and want simple traffic reports. There is a free allowance, then usage-based pricing.","useCases":["See page views and visitor counts","Track traffic without using cookies","Compare traffic across deployments"],"pricing":"Free allowance, then usage-based.","status":"ga","url":"https://vercel.com/docs/analytics","source":"https://vercel.com/docs/products"},{"id":"vercel-speed-insights","provider":"vercel","name":"Speed Insights","category":"observability","tagline":"See how fast your pages feel to real visitors, page by page.","description":"Speed Insights measures how fast your site feels for actual visitors, using Core Web Vitals, Google's scores for loading speed and smoothness. It collects this real-world data for each route, meaning each page address on your site. If you just launched a landing page, you can see whether it feels slow for real people and which pages need work first. That beats a one-time speed test on your own laptop. One older measure, First Input Delay, is deprecated from November 1, 2026.","useCases":["Track how fast real visitors load pages","Monitor Core Web Vitals on each route","Find slow pages by route"],"pricing":"Usage-based.","status":"ga","url":"https://vercel.com/docs/speed-insights","source":"https://vercel.com/docs/products"},{"id":"vercel-observability","provider":"vercel","name":"Vercel Observability","category":"observability","tagline":"See logs, errors and slowdowns for your live Vercel site.","description":"Vercel Observability is a set of dashboards showing what is happening inside your deployed app, including logs (a running record of events), metrics and traces (the step-by-step path of a request). If a customer says checkout is broken, you can open the views for functions, requests and errors to find what failed. You can also search traces from the CLI, the command-line tool. Pick it when you want answers without setting up a separate monitoring product. The basics are included, and an advanced Observability Plus add-on is available.","useCases":["Find the cause of production errors","Watch requests, errors and function metrics","Search request traces from the terminal"],"pricing":"Included, advanced Observability Plus add-on.","status":"ga","url":"https://vercel.com/docs/observability","source":"https://vercel.com/docs/products"},{"id":"vercel-flags","provider":"vercel","name":"Vercel Flags","category":"feature-flags","tagline":"Switch features on or off for chosen users, built into Vercel.","description":"Vercel Flags lets you turn features on or off without rewriting your app, using feature flags, which are simple on/off settings you control. If you are launching a new pricing page, you can show it to a small group of visitors first and widen the audience once it looks stable. It covers flag management and evaluation, with SDK support, meaning ready-made code libraries for your app. Pick it if you already host on Vercel and want flags built into the platform instead of adding another tool. Pricing is usage-based.","useCases":["Release features to a share of users","Turn features on or off","Run gradual rollouts of new changes"],"pricing":"Usage-based.","status":"ga","url":"https://vercel.com/docs/flags","source":"https://vercel.com/docs/products"},{"id":"vercel-image-optimization","provider":"vercel","name":"Image Optimization","category":"images","tagline":"Resize and convert images on demand so your pages load faster.","description":"Image Optimization takes the images you upload and serves each visitor a version sized and formatted for their screen. If your app shows product photos, a phone can get a small file while a large monitor gets a bigger one, so pages load faster. It transforms and caches images, and it is what powers next/image, the image component in Next.js, a popular tool for building websites. Pick it when you want responsive images handled for you rather than resizing files by hand. Pricing is usage-based, covering transformations and cache.","useCases":["Resize images for every screen size","Convert images to lighter file formats","Cache resized images for faster loading"],"pricing":"Usage-based transformations and cache.","status":"ga","url":"https://vercel.com/docs/image-optimization","source":"https://vercel.com/docs/products"},{"id":"vercel-domains","provider":"vercel","name":"Vercel Domains","category":"domains","also":["dns"],"tagline":"Buy and manage domain names, including from the CLI and by AI agents.","description":"Vercel Domains is where you buy a web address like yourbrand.com and manage its DNS, the settings that point the address to your site. If you are launching a new product, you can register the name and connect it to your project in one place. It also works from the CLI, the command-line tool, and by AI agents. Pick it when you want your domain and your hosting managed together, rather than juggling a separate registrar. Prices follow standard registrar pricing for each TLD, the ending such as .com or .io.","useCases":["Buy a domain name for your app","Point DNS records at your projects","Manage domains from the command line"],"pricing":"Registrar pricing per TLD.","status":"ga","url":"https://vercel.com/domains","source":"https://vercel.com/docs/products"},{"id":"vercel-microfrontends","provider":"vercel","name":"Microfrontends","category":"hosting","tagline":"Run several separate apps together so they look like one website.","description":"Microfrontends from Vercel lets you split one website into several smaller apps and show them all under a single web address. Each app can be built and published on its own, and a routing layer (the part that decides which app answers each page) stitches them together. If your marketing site, your dashboard and your docs are built by different teams, visitors still see one site while each team ships on its own schedule. Pick it over cramming everything into one big app when a large team keeps stepping on each other's work, especially in a shared code repository. Pricing is usage-based, so you pay according to how much routing happens.","useCases":["Combine separately deployed apps under one domain","Let big teams deploy apps separately","Route visitors across merged apps"],"pricing":"Usage-based routing.","status":"ga","url":"https://vercel.com/docs/microfrontends","source":"https://vercel.com/docs/products"},{"id":"vercel-agent","provider":"vercel","name":"Vercel Agent","category":"ai-apps","tagline":"An AI helper that reviews your code changes and investigates problems.","description":"Vercel Agent is an AI assistant that reviews your code changes and looks into problems on your live site. It reads pull requests, which are proposed code changes waiting for approval, and analyzes production anomalies, meaning unusual behavior like a sudden spike in errors, right inside the dashboard. If an AI tool wrote most of your code, you can get a second look before merging. When something odd shows up on your live site, it helps you work out where to start. Pricing is credit-based.","useCases":["Review pull requests with AI","Analyze unusual production behavior","Triage incidents inside the dashboard"],"pricing":"Credit-based.","status":"ga","url":"https://vercel.com/docs/agent","source":"https://vercel.com/docs/products"},{"id":"vercel-container-registry","provider":"vercel","name":"Container Registry","category":"devtools","tagline":"A place to store packaged app images, right on Vercel.","description":"Container Registry stores OCI images, which are standard packaged bundles holding an app and everything it needs to run. If you run code in Vercel Sandbox, a safe space for running code, you can keep your prepared environments here and reuse them. The registry accepts pushes, including from GitHub Actions, which is GitHub's tool for running automated tasks. Pick it when you want your images to live next to the rest of your Vercel setup instead of in a separate service. Pricing is usage-based.","useCases":["Store container images for deployments","Push images from GitHub Actions","Keep sandbox images in one place"],"pricing":"Usage-based.","status":"ga","url":"https://vercel.com/docs/container-registry","source":"https://vercel.com/docs/products"},{"id":"vercel-mcp","provider":"vercel","name":"Vercel MCP","category":"agents","tagline":"Let AI agents manage your Vercel projects and read its docs.","description":"Vercel MCP is a hosted server that lets AI agents work with Vercel using MCP, short for Model Context Protocol, a standard way for AI tools to connect to other services. If you use an AI assistant to build your app, it can reach your projects and the docs to help with deploys, so you can ask it to ship a change. It also includes mcp-handler, a helper for building your own MCP servers. Pick it when you want agent-driven deploys without custom wiring. It is free.","useCases":["Let agents deploy and manage projects","Give agents access to project docs","Build custom MCP servers"],"pricing":"Free.","status":"ga","url":"https://vercel.com/docs/mcp","source":"https://vercel.com/docs/products"},{"id":"vercel-connect","provider":"vercel","name":"Vercel Connect","category":"agents","also":["identity"],"tagline":"Connect AI agents to outside services without building each link.","description":"Vercel Connect gives your AI agents managed connectors to third-party services, so you do not build each link yourself. It handles authorization, meaning permission to reach an account, and tool connections for agents. If you are building an assistant that needs to post in Microsoft Teams, this covers that connection, and it also includes TanStack AI and MCP Events. Pick it when agent integrations are slowing you down. It is currently in beta, so expect changes. Pricing details are listed in the docs.","useCases":["Let agents reach third-party services","Manage permissions for agent tools","Connect agents to outside tools"],"pricing":"See docs.","status":"beta","url":"https://vercel.com/changelog","source":"https://vercel.com/docs/products"},{"id":"vercel-drives","provider":"vercel","name":"Vercel Drives","category":"block-storage","tagline":"Storage that stays put for your Vercel Sandbox workspaces.","description":"Vercel Drives are persistent storage volumes you can attach to Vercel Sandbox, a safe space for running code. Normally a temporary workspace starts clean each time. If you have an AI agent that writes code or edits files, a drive lets its workspace survive between sessions, so it picks up where it left off. Pick it when your agent's work should not vanish when a sandbox stops. Drives are now generally available, which means they are ready for production use. Pricing is usage-based.","useCases":["Save agent workspaces to persistent drives","Attach storage volumes to sandboxes","Keep sandbox files on lasting storage"],"pricing":"Usage-based.","status":"ga","url":"https://vercel.com/changelog","source":"https://vercel.com/docs/products"},{"id":"vercel-daemons","provider":"vercel","name":"Vercel Daemons","category":"agents","also":["serverless"],"tagline":"Run background programs that stay on, right on Vercel.","description":"Vercel Daemons are long-running background processes, meaning programs that keep working instead of finishing after a single request. If you are building an always-on agent, such as one that watches for new messages and acts on them, this gives it a home on Vercel. Pick it when your work does not fit the quick in-and-out pattern of a typical web request. It is in beta, with the public beta announced October 15, 2026. Check the changelog for pricing.","useCases":["Run always-on background processes","Host long-running agents","Keep background jobs running continuously"],"pricing":"See changelog.","status":"beta","url":"https://vercel.com/changelog","source":"https://vercel.com/docs/products"},{"id":"vercel-eve","provider":"vercel","name":"Eve","category":"agents","tagline":"Free, open source toolkit for building AI agents that remember things.","description":"Eve is a free, open source framework, meaning a ready-made toolkit whose code you can read and change, for building AI agents. An agent is an AI program that carries out tasks on its own instead of only answering questions. If you want a support helper that remembers what each customer asked last week, Eve's persistent memory covers that. You can deploy what you build straight from the Vercel dashboard. Pick it when you want a complete agent setup you control, with no license fee.","useCases":["Build agents with persistent memory","Deploy agent apps from the dashboard","Start from an open source agent framework"],"pricing":"Free, open source.","status":"ga","url":"https://vercel.com/platform","source":"https://vercel.com/docs/products"},{"id":"vercel-secure-compute","provider":"vercel","name":"Secure Compute","category":"networking","tagline":"A private network lane for your Vercel app, with fixed IP addresses.","description":"Secure Compute puts your builds and functions (small pieces of code that run on demand) on an isolated private network with dedicated IP addresses, meaning fixed addresses other systems can recognize. If your app needs to read from a database that only accepts connections from approved addresses, this is how you reach it. It also supports VPC peering and AWS PrivateLink, which are ways to link privately to your AWS network. Pick it over the default setup when locked-down database access is a requirement. It is an Enterprise add-on.","useCases":["Reach private databases from builds","Use static IPs for allowlists","Connect through VPC peering or PrivateLink"],"pricing":"Enterprise add-on.","status":"ga","url":"https://vercel.com/docs/secure-compute","source":"https://vercel.com/docs/products"},{"id":"vercel-byoc-aws","provider":"vercel","name":"Bring Your Own Cloud on AWS","category":"hosting","tagline":"Run Vercel inside your own AWS account.","description":"Bring Your Own Cloud on AWS lets you run Vercel in your own Amazon Web Services account instead of Vercel's. If your customers require their data to stay in a certain region, or your company already has a spending commitment with AWS, this keeps both in line. You get Vercel's hosting, but inside an account you own. Pick it over standard Vercel hosting when contracts or rules decide where your app has to run. It has been generally available since October 2026 and is an Enterprise offering.","useCases":["Run Vercel inside your AWS account","Meet data residency requirements","Use existing AWS commitments"],"pricing":"Enterprise.","status":"ga","url":"https://vercel.com/changelog","source":"https://vercel.com/docs/products"},{"id":"vercel-cron-jobs","provider":"vercel","name":"Vercel Cron Jobs","category":"workflows","tagline":"Run a task on a schedule, like every morning at 8.","description":"Vercel Cron Jobs call a web address in your app on a schedule you set. You write the schedule in a config file using cron expressions, a short code format where 0 8 * * * means every day at 8. If you want a daily report built and sent to your team, a cron job can trigger the code that does it. Pick it when you want timed tasks living next to your app, with no separate scheduler to set up. It is included, with limits depending on your plan.","useCases":["Run tasks on a daily schedule","Send daily reports automatically","Trigger web endpoints on a schedule"],"pricing":"Included, limits by plan.","status":"ga","url":"https://vercel.com/docs/cron-jobs","source":"https://vercel.com/docs/products"},{"id":"vercel-toolbar","provider":"vercel","name":"Vercel Toolbar","category":"devtools","tagline":"Leave comments and inspect your preview sites with a built-in overlay.","description":"The Vercel Toolbar is a small overlay on your preview deployments, the test versions of your site you see before going live. Teammates can leave comments right on the page, which makes design review much easier. It also includes feature flags (switches that turn features on or off) and tools to spot layout shift, which is when things jump around as a page loads. Pick it over screenshots and chat threads when you want feedback tied to the exact spot on the page. It is free.","useCases":["Leave feedback on preview deployments","Inspect preview deployments up close","Check layout shifts on pages"],"pricing":"Free.","status":"ga","url":"https://vercel.com/docs/vercel-toolbar","source":"https://vercel.com/docs/products"},{"id":"netlify-hosting","provider":"netlify","name":"Netlify Hosting and CDN","category":"hosting","tagline":"Publish your site from Git, with global delivery and preview links.","description":"Netlify Hosting and CDN puts your website online each time you push code to Git, the tool that tracks code changes, and serves it through a CDN, a network that delivers files from a location near each visitor. If you are building a marketing site or an app, every branch gets its own preview link to share before launch. Deploys are atomic, so visitors never see a half-updated site, and you can roll back to an earlier version. Plans are now credit-based, and a free plan is available.","useCases":["Deploy a site from Git","Preview each change at its own URL","Roll back a bad release"],"pricing":"Credit-based. Free plan available.","status":"ga","url":"https://www.netlify.com/platform/","source":"https://www.netlify.com/platform/"},{"id":"netlify-functions","provider":"netlify","name":"Netlify Functions","category":"serverless","tagline":"Small pieces of backend code that run on demand, deployed with your site.","description":"Netlify Functions let you run backend code without managing a server. They can respond to requests, run in the background, or run on a schedule. If your site has a contact form that needs to pass data to another service, or you need to receive webhooks (automatic messages sent by other services), a function handles it. You write them in JavaScript, TypeScript or Go, and they deploy alongside your site. Pick them when you want a simple API next to the part visitors see, without a separate backend project. Usage is credit-based.","useCases":["Run API endpoints alongside your site","Handle incoming webhooks from other services","Run scheduled tasks on a timer"],"pricing":"Credit-based.","status":"ga","url":"https://docs.netlify.com/functions/overview/","source":"https://www.netlify.com/platform/"},{"id":"netlify-edge-functions","provider":"netlify","name":"Netlify Edge Functions","category":"edge-compute","tagline":"Code that runs near each visitor, for things like location-based changes.","description":"Netlify Edge Functions run small bits of code on servers close to your visitors, before the page reaches them. They use Deno, a modern runtime for JavaScript and TypeScript. If you want to show prices in a local currency or send half your visitors to a different headline for an A/B test, this is the place to do it. They also handle rewrites, which quietly serve a different page at the same address. Pick them over regular Netlify Functions when per-visitor changes matter most. Usage is credit-based.","useCases":["Personalize pages for each visitor","Show content based on visitor location","Split traffic for A/B tests"],"pricing":"Credit-based.","status":"ga","url":"https://docs.netlify.com/build/edge-functions/overview/","source":"https://www.netlify.com/platform/"},{"id":"netlify-async-workloads","provider":"netlify","name":"Async Workloads","category":"workflows","also":["queues"],"tagline":"Background jobs that retry on failure, with queues and multi-step flows.","description":"Async Workloads run work in the background so your site does not make visitors wait. Jobs are durable, meaning they retry if something fails, and they come with built-in queues and delays. If your app sends a welcome email after signup, or runs a long job like processing a big file, you can hand it off and move on. Multi-step workflows let you chain tasks in order. Pick it over a plain function when a job must finish even if something goes wrong along the way. Usage is credit-based.","useCases":["Send emails without slowing pages","Run long jobs in the background","Retry failed work automatically"],"pricing":"Credit-based.","status":"ga","url":"https://docs.netlify.com/","source":"https://www.netlify.com/platform/"},{"id":"netlify-blobs","provider":"netlify","name":"Netlify Blobs","category":"object-storage","also":["kv-cache"],"tagline":"Save files and simple data with no setup, right from your code.","description":"Netlify Blobs is built-in storage for files and structured data, with nothing to set up. If your app lets people upload files, or you want to cache results so pages load faster, you can save them from a function or a build. It works as object storage, which holds files, and as key-value storage, where you look up data by a name. Pick it over a separate storage service when you want the simplest option that already lives inside Netlify. Usage is credit-based.","useCases":["Cache data for faster responses","Store user files from functions","Save structured data without setup"],"pricing":"Credit-based.","status":"ga","url":"https://docs.netlify.com/build/data-and-storage/netlify-blobs/","source":"https://www.netlify.com/platform/"},{"id":"netlify-database","provider":"netlify","name":"Netlify Database","category":"sql","tagline":"Postgres database set up with your deploy, with a branch per preview.","description":"Netlify Database is serverless Postgres, a popular database, meaning you do not manage the server behind it. It is created when you deploy, and each deploy preview gets its own branch, a separate copy where you can test changes safely. If you are building an app that stores users and orders, or an AI feature that searches your own documents using pgvector (an add-on for searching by meaning), it fits well. It launched April 28, 2026, replacing the beta Netlify DB. Compute and bandwidth use credits.","useCases":["Store app data in Postgres","Give each preview its own database branch","Power AI search with vector data"],"pricing":"Credit-based; compute and bandwidth consume credits.","status":"ga","url":"https://docs.netlify.com/build/data-and-storage/netlify-database/","source":"https://www.netlify.com/platform/"},{"id":"netlify-image-cdn","provider":"netlify","name":"Netlify Image CDN","category":"images","tagline":"Resize and convert images on the fly just by changing the link.","description":"Netlify Image CDN changes your images as they are requested, using options added to the image link. If your site shows photos on both phones and big monitors, you can ask for a smaller version for each instead of making many copies by hand. It resizes images and converts their format, which helps pages load faster. Pick it when you want responsive images, meaning pictures that fit each screen, without adding a separate image tool to your setup. Usage is credit-based.","useCases":["Resize images through URL settings","Convert images to other formats","Serve responsive images for each screen"],"pricing":"Credit-based.","status":"ga","url":"https://docs.netlify.com/build/image-cdn/overview/","source":"https://www.netlify.com/platform/"},{"id":"netlify-forms","provider":"netlify","name":"Netlify Forms","category":"devtools","tagline":"Collect form submissions on your site without writing server code.","description":"Netlify Forms lets your website collect what people type into a form, such as a name and a message, without you building a back end (the behind-the-scenes part of an app). If your site has a contact form, Netlify spots the form when your site is built and saves every submission for you, with spam filtering included. That makes it a quick pick over setting up your own database or a separate form service for simple needs. You get a free allowance, then it uses credits.","useCases":["Collect contact form submissions","Add forms without backend code","Filter spam from form submissions"],"pricing":"Free allowance, then credits.","status":"ga","url":"https://docs.netlify.com/forms/setup/","source":"https://www.netlify.com/platform/"},{"id":"netlify-identity","provider":"netlify","name":"Netlify Identity","category":"identity","tagline":"Add sign-up and login to your site with simple user management.","description":"Netlify Identity handles the sign-up and login part of a site, so people can create accounts and come back later. If you are building a members-only page or a simple app with user accounts, it covers basic registration and login without you building it from scratch. Netlify once planned to retire it but reversed that in February 2026, so it is still supported. For new projects, Netlify points you to @netlify/identity, a code library made for this job. Identity is included on credit-based plans.","useCases":["Let people register and log in","Manage your users in one place","Add basic login to a site"],"pricing":"Included on credit-based plans.","status":"ga","url":"https://docs.netlify.com/security/secure-access-to-sites/identity/","source":"https://www.netlify.com/platform/"},{"id":"netlify-ai-gateway","provider":"netlify","name":"Netlify AI Gateway","category":"ai-gateway","also":["ai-models"],"tagline":"Use major AI models in your site without managing API keys.","description":"Netlify AI Gateway gives your site access to major AI models without you signing up with each AI company and pasting in API keys (secret codes that identify you to a service). If you want to add a chat box or a text summarizer to your site, you can call a model straight from your Netlify project. Because access is built into the platform, usage is billed through your Netlify credits instead of separate accounts. That is handy when you want one bill and less setup.","useCases":["Add AI models with no API keys","Call major AI models from your site","Pay for AI through Netlify credits"],"pricing":"Credit-based.","status":"ga","url":"https://www.netlify.com/platform/ai-gateway","source":"https://www.netlify.com/platform/"},{"id":"netlify-agent-runners","provider":"netlify","name":"Netlify Agent Runners","category":"agents","tagline":"Have an AI coding agent build or fix your project from the dashboard.","description":"Netlify Agent Runners let you pick an AI coding agent, such as Claude, Gemini, Codex or OpenCode, and have it create or edit your project from the Netlify dashboard. If you have an idea, you can describe it in a prompt and get a working draft, or ask for a bug fix from your phone. The agents have access to your deploy context, and each change produces a Deploy Preview, a private test link you can check before anything goes live. It is available on credit-based plans only.","useCases":["Build a project from a prompt","Fix bugs from your phone","Get a Deploy Preview for each change"],"pricing":"Credit-based plans only.","status":"ga","url":"https://www.netlify.com/platform/agent-runners","source":"https://www.netlify.com/platform/"},{"id":"netlify-analytics","provider":"netlify","name":"Netlify Analytics","category":"analytics","tagline":"See who visits your site, counted from server logs with no tracking script.","description":"Netlify Analytics shows you how much traffic your site gets. Instead of adding a tracking script to your pages, it reads the logs from Netlify's CDN (the network of servers that delivers your site to visitors). If you want simple traffic reports without adding extra code to your site, this is an easy fit, since there is nothing for your visitors' browsers to load. It is an add-on, so check the Netlify site for current pricing.","useCases":["See traffic reports for your site","Measure visits without client scripts","Track traffic from server logs"],"pricing":"Add-on pricing. See site.","status":"ga","url":"https://docs.netlify.com/manage/monitoring/web-analytics/overview/","source":"https://www.netlify.com/platform/"},{"id":"netlify-dns","provider":"netlify","name":"Netlify DNS","category":"dns","tagline":"Manage your domain's settings in Netlify, with records set up for you.","description":"Netlify DNS manages the address book that tells the internet where your domain lives. DNS, short for Domain Name System, is what turns a name like yourstore.com into the server that shows your site. If your site is hosted on Netlify, you can hand your domain over so the records for your deploys are created automatically. That saves you from editing settings by hand each time something changes. It is included with Netlify.","useCases":["Manage DNS records for your domain","Host DNS zones on Netlify","Get records set up on deploy"],"pricing":"Included.","status":"ga","url":"https://docs.netlify.com/manage/domains/set-up-netlify-dns/","source":"https://www.netlify.com/platform/"},{"id":"netlify-build","provider":"netlify","name":"Netlify Build and Deploy Previews","category":"devtools","tagline":"Automatic builds from your code, with a preview link for every change.","description":"Netlify Build and Deploy Previews turns your code into a live site every time you push a change in Git (the tool that tracks versions of your code). If a teammate or an AI tool opens a pull request, which is a proposed change, you get a preview link to see how the site will look before it goes live. Your team can leave comments right on the preview. Build plugins let you add extra steps to the process. Pricing is credit-based.","useCases":["Get a preview URL per pull request","Review changes with team comments","Run builds automatically from Git"],"pricing":"Credit-based.","status":"ga","url":"https://docs.netlify.com/build/configure-builds/overview/","source":"https://www.netlify.com/platform/"},{"id":"netlify-monitoring","provider":"netlify","name":"Monitoring and Insights","category":"observability","tagline":"Check logs and traffic stats to find out why your site or functions break.","description":"Monitoring and Insights helps you see what is happening behind your site. If a function, a small piece of back-end code, fails when someone submits a form, you can read its logs to find out why. You also get traffic stats, plus log drains, which send your logs to outside tools you already use. It is useful once your site has real visitors and you need answers fast. What is included depends on your plan.","useCases":["Read function logs when debugging","Send logs to outside tools","Check traffic stats and telemetry"],"pricing":"Plan dependent.","status":"ga","url":"https://docs.netlify.com/manage/monitoring/logs/","source":"https://www.netlify.com/platform/"},{"id":"digitalocean-droplets","provider":"digitalocean","name":"Droplets","category":"vms","tagline":"Rent a Linux server in the cloud, starting at $4 a month.","description":"Droplets are DigitalOcean's basic cloud servers: virtual machines running Linux that you rent by the month. If you need to run a web server or set up a development space to test things, a Droplet gives you a whole machine to control. You can choose shared or dedicated CPU, the part that does the computing, with prices from $4 a month. Pick it over a managed hosting service when you want full control of the setup. DigitalOcean also announced Agent Droplets in October 2026, which bundle lightweight virtual machines with AI model access and storage for AI agents.","useCases":["Run web servers on virtual machines","Set up development environments","Spin up Linux virtual machines"],"pricing":"From $4/month.","status":"ga","url":"https://www.digitalocean.com/products/droplets","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-gpu-droplets","provider":"digitalocean","name":"GPU Droplets","category":"gpu","tagline":"Rent graphics-chip servers by the hour to train or run AI models.","description":"GPU Droplets are cloud servers with NVIDIA or AMD GPUs, the graphics chips that do the heavy math behind AI. If you want to train your own AI model or run one for your app's users, you can start a server on demand and pay by the hour per GPU. Options include H100, H200 and MI300X chips, and they come with AI-ready setups and 1-Click Models, ready-made models you can start with one click. Pick this when you need AI hardware without a long contract.","useCases":["Train AI models on GPUs","Run AI model inference","Launch GPU machines by the hour"],"pricing":"Hourly per GPU.","status":"ga","url":"https://www.digitalocean.com/products/gpu-droplets","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-bare-metal-gpus","provider":"digitalocean","name":"Bare Metal GPUs","category":"gpu","tagline":"Whole servers with 8 GPUs, reserved for you alone, for big AI jobs.","description":"Bare Metal GPUs are dedicated servers with 8 GPUs each, and nobody else shares the machine. If you are training a large AI model or serving one to many users, you get the full speed of the hardware without sharing it. You can use one server on its own or link several into a cluster, a group of machines working together. Pick this over GPU Droplets when your job is big enough to need whole machines. Pricing is by contract or hourly, so check DigitalOcean's site.","useCases":["Run large training jobs","Use dedicated 8-GPU servers","Scale across GPU clusters"],"pricing":"Contract or hourly. See site.","status":"ga","url":"https://www.digitalocean.com/products/bare-metal-gpu","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-app-platform","provider":"digitalocean","name":"App Platform","category":"hosting","also":["containers"],"tagline":"Put your app online without managing servers, straight from your code.","description":"App Platform is DigitalOcean's way to put your app online without looking after servers yourself. If you have a web app, a static site or a background worker (code that does jobs out of sight), you connect your Git repository or a container, which is a packaged app, and it deploys. It scales up with traffic automatically and handles the security certificates that give your site HTTPS. Pick it over a Droplet when you would rather not set up and maintain a server. Static sites are free, and containers start at $5 a month.","useCases":["Deploy web apps from Git","Run background workers and jobs","Scale apps automatically"],"pricing":"Static sites free. Containers from $5/month.","status":"ga","url":"https://www.digitalocean.com/products/app-platform","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-functions","provider":"digitalocean","name":"Functions","category":"serverless","tagline":"Run small bits of code on demand, without managing any servers.","description":"Functions lets you run a small piece of code only when something triggers it, with no server for you to set up or look after. If a payment service needs to notify your app when a customer pays (this is called a webhook, an automatic message from one service to another), a function can catch it and update your records. It also suits light APIs, the behind-the-scenes addresses your app calls to get data. Pick it over an always-on server when your traffic is uneven, since it scales automatically and is priced by usage, with a free monthly allowance.","useCases":["Run webhook handlers on demand","Serve light API endpoints","Scale functions with incoming events"],"pricing":"Usage-based with a free monthly allowance.","status":"ga","url":"https://www.digitalocean.com/products/functions","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-kubernetes","provider":"digitalocean","name":"DigitalOcean Kubernetes (DOKS)","category":"containers","tagline":"Run containerized apps across many servers, with the setup handled.","description":"DigitalOcean Kubernetes (DOKS) is a managed way to run Kubernetes, a popular system for running apps packed in containers (portable bundles of code) across a group of servers. If your product is split into several small services, often called microservices, it keeps them running and organized. It also supports GPU node pools, which are groups of servers with graphics chips, for AI and other heavy workloads. The management layer, called the control plane, is free. You pay for the worker servers, and a high-availability version of the control plane costs extra.","useCases":["Run microservices in containers","Manage GPU workloads in clusters","Use CPU and GPU node pools"],"pricing":"Pay for worker nodes. HA control plane extra.","status":"ga","url":"https://www.digitalocean.com/products/kubernetes","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-container-registry","provider":"digitalocean","name":"Container Registry","category":"containers","tagline":"A private place to store your app's Docker images.","description":"Container Registry is a private shelf for Docker images, the packaged versions of your app that get deployed and run. If you bundle your app into a container, you push the image here and your hosting pulls it when it is time to launch. It is integrated with DOKS (DigitalOcean's Kubernetes service) and App Platform, so it fits neatly with both. Pick it over a public registry when you do not want your code visible to everyone. There is a free tier, and paid plans start at $5/month.","useCases":["Store Docker images privately","Pull images into Kubernetes clusters","Deploy images to App Platform"],"pricing":"Free tier. Paid from $5/month.","status":"ga","url":"https://www.digitalocean.com/products/container-registry","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-postgresql","provider":"digitalocean","name":"Managed PostgreSQL","category":"sql","tagline":"A ready-to-use Postgres database with backups and backup servers.","description":"Managed PostgreSQL gives you Postgres, a popular database, that DigitalOcean sets up and looks after for you. If your app needs to store users, orders or posts, this is where they would live. It includes standby nodes (spare copies that take over if the main one fails), read replicas (extra copies that handle lookups), and PgBouncer, a tool that manages many connections at once. Pick it over running a database yourself when you would rather not handle upkeep. Plans start at $15/month.","useCases":["Host app databases with backups","Keep databases running with standby nodes","Add read replicas for busy apps"],"pricing":"From $15/month.","status":"ga","url":"https://www.digitalocean.com/products/managed-databases-postgresql","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-mysql","provider":"digitalocean","name":"Managed MySQL","category":"sql","tagline":"A ready-to-use MySQL database with automatic backups.","description":"Managed MySQL gives you MySQL, one of the most widely used databases, that DigitalOcean sets up and maintains for you. If you are running a web app or a site that stores customers, products or content, this is a natural home for that data. It takes automated backups, switches to a standby copy if the main one fails, and offers read-only nodes (extra copies that handle lookups). Pick it over running MySQL on your own server when you want fewer chores. Plans start at $15/month.","useCases":["Host web app data in MySQL","Keep data safe with automated backups","Fail over to standby nodes"],"pricing":"From $15/month.","status":"ga","url":"https://www.digitalocean.com/products/managed-databases-mysql","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-valkey","provider":"digitalocean","name":"Managed Valkey","category":"kv-cache","tagline":"A fast in-memory store for caching and queues, Redis-compatible.","description":"Managed Valkey is a fast data store that keeps information in memory, which makes it great for things your app needs again and again. If your pages load slowly because they keep asking the main database the same question, you can cache (save) the answers here. It also works for queues, which are lists of jobs waiting to run. It is compatible with Redis, a well-known tool of the same kind, and it replaced DigitalOcean's Managed Caching (Redis) product as Redis licensing changed. Plans start at $15/month.","useCases":["Cache data for faster responses","Run job queues for background work","Speed up repeated database lookups"],"pricing":"From $15/month.","status":"ga","url":"https://www.digitalocean.com/products/managed-databases-valkey","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-mongodb","provider":"digitalocean","name":"Managed MongoDB","category":"nosql","tagline":"A hosted MongoDB database for flexible, document-style data.","description":"Managed MongoDB runs MongoDB, a database that stores information as flexible documents instead of rigid tables, and DigitalOcean handles the setup. If your app saves things like user profiles or product listings where each item can have different details, this is a natural fit. Clusters are replica sets (several copies of your data kept in sync), with backups and scaling included. Pick it over a table-based database when the shape of your data changes often. Plans start at $15/month.","useCases":["Store app data as documents","Run replicated databases with backups","Scale database capacity as traffic grows"],"pricing":"From $15/month.","status":"ga","url":"https://www.digitalocean.com/products/managed-databases-mongodb","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-kafka","provider":"digitalocean","name":"Managed Kafka","category":"queues","tagline":"Handle a constant stream of events between parts of your app.","description":"Managed Kafka runs Apache Kafka, a system for moving a constant flow of events, like clicks, orders or sensor readings, between different parts of your software. If every purchase should update inventory, send an email and feed your analytics at once, you publish it once to a topic (a named channel) and each service picks it up. It includes ACLs (rules about who can read or write) and monitoring. Pick it when you have heavy streaming needs, since it is a bigger tool than a simple job queue. Plans start at $147/month.","useCases":["Stream live events between services","Stream user activity as it happens","Monitor event streams and who accesses them"],"pricing":"From $147/month.","status":"ga","url":"https://www.digitalocean.com/products/managed-databases-kafka","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-opensearch","provider":"digitalocean","name":"Managed OpenSearch","category":"nosql","also":["observability"],"tagline":"Add search to your site or dig through logs, hosted for you.","description":"Managed OpenSearch is a hosted search and analytics engine, a tool built to find things quickly in large piles of text. If your store needs a search box that returns products fast, or you want to comb through your app's logs (records of what happened) to track down errors, this fits both jobs. DigitalOcean runs the clusters for you, so there is nothing to install on your own servers. Pick it over a regular database search when you need quick, flexible text lookup. Plans start at $15/month.","useCases":["Power search boxes on your site","Search through application logs","Run analytics across large datasets"],"pricing":"From $15/month.","status":"ga","url":"https://www.digitalocean.com/products/managed-databases-opensearch","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-spaces","provider":"digitalocean","name":"Spaces Object Storage","category":"object-storage","tagline":"Store your app's files and images, with a built-in CDN.","description":"Spaces Object Storage is online storage for files such as images, videos, backups and website assets. If your app lets people upload photos, you keep them here instead of on your own server. It is S3-compatible, meaning tools built for Amazon's S3 storage work with it, and it has a built-in CDN, a network that delivers files quickly to visitors around the world. The flat $5/month plan covers 250 GB of storage and 1 TB of transfer, which makes costs easy to predict.","useCases":["Store uploaded files in buckets","Serve images and downloads quickly","Keep off-site backup copies"],"pricing":"$5/month for 250 GB and 1 TB transfer.","status":"ga","url":"https://www.digitalocean.com/products/spaces","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-volumes","provider":"digitalocean","name":"Volumes Block Storage","category":"block-storage","tagline":"Extra disk space you can attach to your server and grow later.","description":"Volumes Block Storage is extra disk space that you attach to a Droplet, which is DigitalOcean's name for a server. If your database is running out of room, you add a volume and keep going. The volumes use NVMe, a fast type of storage, and can be resized without downtime, so your app stays online while you grow. Pick it over a bigger server when you only need more space, not more computing power. Pricing is $0.10 per GB per month.","useCases":["Attach extra disk space to servers","Hold database files on fast disks","Grow disk size without downtime"],"pricing":"$0.10/GB-month.","status":"ga","url":"https://www.digitalocean.com/products/block-storage","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-nfs","provider":"digitalocean","name":"Network File Storage","category":"block-storage","tagline":"A shared folder that several servers can use at the same time.","description":"Network File Storage is a shared file system, like a folder in the cloud that several machines can open at once. If your app runs on more than one server and they all need the same uploaded media, this keeps everyone looking at the same files. It uses NFS, a standard way of sharing folders over a network, and can be mounted by multiple Droplets (DigitalOcean servers) and Kubernetes clusters. Pick it over Volumes when more than one machine needs the same files together. Pricing is based on usage.","useCases":["Share one folder across several servers","Let multiple machines read the same files","Hold shared media libraries"],"pricing":"Usage-based.","status":"ga","url":"https://docs.digitalocean.com/products/nfs/","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-backups","provider":"digitalocean","name":"Backups and Snapshots","category":"block-storage","tagline":"Automatic copies of your server, so you can restore if something breaks.","description":"Backups and Snapshots saves copies of your Droplet, which is DigitalOcean's name for a server. Backups run automatically on a weekly or daily schedule, and you can also take a snapshot, a saved copy of your server at one moment, whenever you want. If you are about to ship a risky update, take a snapshot first so you can roll back if it goes wrong. Weekly backups cost 20% of your Droplet's price, which makes them an inexpensive safety net.","useCases":["Back up servers automatically on schedule","Save snapshots of servers","Restore a server from a saved image"],"pricing":"20% of Droplet price for weekly backups.","status":"ga","url":"https://www.digitalocean.com/products/backups","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-load-balancers","provider":"digitalocean","name":"Load Balancers","category":"load-balancing","tagline":"Spread visitors across several servers so your site stays up under load.","description":"A load balancer is a traffic director that spreads visitors across several servers. If your app grows past what one server can handle, you put a load balancer in front of two or more servers and it sends each visitor to one that is healthy, using regular health checks. It also handles TLS, the encryption behind the browser padlock, and works at Layer 4 and Layer 7, meaning it can route basic connections or look inside web requests. DigitalOcean manages it for you within a region. Pricing starts at $12/month.","useCases":["Spread traffic across several web servers","Check server health and route around failures","Handle HTTPS encryption at the edge"],"pricing":"From $12/month.","status":"ga","url":"https://www.digitalocean.com/products/load-balancer","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-dns","provider":"digitalocean","name":"DigitalOcean DNS","category":"dns","tagline":"Free hosting for your domain's address book, so your name points to your app.","description":"DNS is the address book that tells browsers which server a domain name like yourapp.com points to. DigitalOcean DNS hosts that address book for you at no cost. If you just bought a domain and want it to open your app, you add it here and create the records that point it to your server. It also has an API, a way for code to make changes automatically, which helps if you manage many domains or want to script updates.","useCases":["Point your domain to your servers","Manage domain records through an API","Host DNS records at no cost"],"pricing":"Free.","status":"ga","url":"https://docs.digitalocean.com/products/networking/dns/","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-vpc","provider":"digitalocean","name":"VPC Networking and Firewalls","category":"networking","tagline":"Private networks and free firewalls to keep your servers locked down.","description":"This is the set of tools for controlling who can reach your servers. A VPC, or virtual private cloud, is a private network where your resources talk to each other without being exposed to the public internet. If your database should only be reachable by your web server, put both in a VPC and add a Cloud Firewall, which blocks other traffic and is free. Reserved IPs (fixed addresses), NAT gateways (which let private servers reach out to the internet) and links to outside networks are included too. Most of it is free, while NAT and partner links cost extra.","useCases":["Keep internal servers off the public internet","Set firewall rules for your servers","Give private servers outbound internet access"],"pricing":"Mostly free. NAT and partner links extra.","status":"ga","url":"https://www.digitalocean.com/products/vpc","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-uptime","provider":"digitalocean","name":"Uptime and Monitoring","category":"observability","tagline":"Get alerted when your site goes down or your server is struggling.","description":"Uptime and Monitoring watches your site and your servers and tells you when something looks wrong. If you launch a landing page or app and want to hear about an outage before your customers do, you set up an uptime check, which visits your address regularly and alerts you if it fails. It also tracks health numbers for your Droplets, which are DigitalOcean's servers, with alerts. It is basic monitoring for DigitalOcean resources, so pick it when you want simple coverage without adding another tool. Free tiers are available.","useCases":["Get alerted when your site goes down","Check site availability from outside","Track metrics for your servers"],"pricing":"Free tiers; see pricing.","status":"ga","url":"https://www.digitalocean.com/products/uptime-monitoring","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-cspm","provider":"digitalocean","name":"Cloud Security Posture Management","category":"threat-intel","tagline":"Scans your DigitalOcean account for risky settings and mistakes.","description":"Cloud Security Posture Management checks your DigitalOcean account for settings that could leave you exposed. If you have been moving fast and spinning up servers and databases, a scan can find misconfigurations, meaning things set up in a risky or unintended way. It looks across the whole account rather than one resource at a time. It is a good fit when you want a regular checkup without hiring a security specialist. Check the pricing page for cost.","useCases":["Find risky settings across your account","Audit cloud setup for misconfigurations","Spot exposed resources before attackers do"],"pricing":"See pricing.","status":"ga","url":"https://www.digitalocean.com/products","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-iam","provider":"digitalocean","name":"Identity and Access Management","category":"identity","tagline":"Decide who on your team can see and change what.","description":"Identity and Access Management lets you control what each person on your team can do in your DigitalOcean account. If you bring on a freelancer to work on your website, you can give them access to only the resources they need instead of handing over the keys to everything. Permissions are fine-grained, meaning you can set limits at a detailed level across your resources. It is free, so there is little reason to skip it once more than one person is involved.","useCases":["Give teammates limited access rights","Control who can change what","Set permissions for each resource"],"pricing":"Free.","status":"ga","url":"https://www.digitalocean.com/products","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-agent-platform","provider":"digitalocean","name":"Gradient AI Agent Platform","category":"ai-platform","also":["agents"],"tagline":"Build AI assistants that answer from your own documents.","description":"Gradient AI Agent Platform is a managed way to build AI agents, which are assistants that answer questions and take actions. If you want a support bot that answers from your help docs, you add them to a knowledge base, which lets the AI look up your content before replying (a technique called RAG). It also offers guardrails, routing, function routes and evaluation tools to check how well your agent works. It was previously called the GenAI Platform or Gradient AI Platform. The Gradient AI SDK was deprecated on July 15, 2026 in favor of standard client libraries.","useCases":["Build assistants that search your documents","Connect agents to your own knowledge bases","Coordinate several agents in one app"],"pricing":"Usage-based.","status":"ga","url":"https://www.digitalocean.com/products/gradient/platform","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-serverless-inference","provider":"digitalocean","name":"Serverless Inference","category":"ai-models","tagline":"One connection to 70+ AI models for chat, images, audio and more.","description":"Serverless Inference lets your app use AI models through a single API, without you running any hardware. If you want to add a chat feature or generate images in your product, you send a request to one endpoint and get a response back. It gives you access to more than 70 open and commercial models, covering chat, images, audio, embeddings (numbers that help search by meaning) and responses APIs. You pay per token, which is a small chunk of text, so you only pay for what you use. Pick it to try different models without managing servers.","useCases":["Add AI text features with one API","Generate images, audio or embeddings","Pay per token for model calls"],"pricing":"Per token.","status":"ga","url":"https://www.digitalocean.com/products/inference-engine","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-dedicated-inference","provider":"digitalocean","name":"Dedicated Inference","category":"ai-models","tagline":"Reserve your own GPU for running AI models in production.","description":"Dedicated Inference gives you your own GPU endpoints, meaning reserved graphics chips that run an AI model only for you. If your app sends a steady stream of requests to a language model, a dedicated setup can suit you better than sharing capacity. You choose the accelerator type and size, and the GPU is billed hourly. Serverless Inference is the easier choice for light or unpredictable use. Dedicated Inference is currently in public preview, so expect some changes.","useCases":["Host a model on reserved GPUs","Serve steady high-volume AI traffic","Pick GPU type and size"],"pricing":"Hourly GPU.","status":"preview","url":"https://docs.digitalocean.com/products/inference/","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-batch-inference","provider":"digitalocean","name":"Batch Inference","category":"ai-models","tagline":"Send a big pile of AI requests as one job, at a lower price.","description":"Batch Inference lets you submit many AI requests together as a single job instead of sending them one at a time. The answers come back later rather than instantly. If you need to summarize thousands of support tickets or tag a whole product catalog overnight, this fits well. Because nobody is waiting on each response, you get a discounted price per token, which is a small chunk of text. Pick it for offline work, and use Serverless Inference when your users need an instant reply.","useCases":["Run big batches of AI requests","Process content without waiting for replies","Save money on bulk AI jobs"],"pricing":"Discounted per token.","status":"ga","url":"https://docs.digitalocean.com/products/inference/","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-inference-router","provider":"digitalocean","name":"Inference Router","category":"ai-gateway","also":["ai-models"],"tagline":"Let one router pick the right AI model for each request.","description":"Inference Router is a single managed endpoint that chooses which AI model handles each request. If your app asks AI both simple and tricky questions, the router can send easy ones to a cheaper model and hard ones to a stronger one, which helps cut costs. It was redesigned on October 12, 2026 with balanced, budget and quality preferences, so you tell it what matters most. You pay for the usage of the models it routes to. It is currently in public preview.","useCases":["Pick a model for each request automatically","Trade off cost against answer quality","Lower AI bills by routing to cheaper models"],"pricing":"Routed model usage.","status":"preview","url":"https://docs.digitalocean.com/products/inference/","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-managed-agents","provider":"digitalocean","name":"Managed Agents and Agent Droplets","category":"agents","tagline":"A home for your AI agents with storage, AI models and tools in one monthly plan","description":"Managed Agents and Agent Droplets give AI agents, programs that do tasks on their own, a walled-off place to run, with storage, AI model access and tools in one plan. If you want to run Claude Code or Codex agents for your product, you can start them here instead of stitching together a server, storage and model account. Each one runs in a microVM, a small isolated computer, and Action Gateway gives it access to more than 16,000 tools under rules you set. Choose it when you would rather pay one monthly Agent Droplet plan than juggle several separate bills.","useCases":["Run coding agents in isolated environments","Host agent runtimes with tools and storage","Give agents access to many tools"],"pricing":"Monthly Agent Droplet plans.","status":"ga","url":"https://www.digitalocean.com/blog/introducing-agent-droplets","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-evaluations","provider":"digitalocean","name":"Learning Services (Evaluations, Signals)","category":"ai-platform","tagline":"Repeatable tests that check whether your AI agent still answers well.","description":"Learning Services is a set of tools for checking how good your AI agents are. Evaluations are repeatable tests where one AI model acts as the judge (called LLM-as-judge) and grades your agent's answers. If you change a prompt or swap to a different model, you can rerun the same tests to catch anything that got worse, which is known as regression testing. Signals and Simulations are also part of it, and both are in public preview. Pick it when you want a steady way to measure agent quality instead of trying a few chats by hand.","useCases":["Test agent answers after every change","Catch quality drops before users notice","Score model outputs with automated judges"],"pricing":"See pricing.","status":"preview","url":"https://www.digitalocean.com/products","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-content-safety","provider":"digitalocean","name":"Content Safety Guardrails","category":"ai-gateway","also":["ai-models"],"tagline":"Screens what your AI reads and says, then allows, flags or blocks it.","description":"Content Safety Guardrails are built-in checks that sit around the calls your app makes to an AI model. They look at what users send in and what the model sends back, then decide to allow it, flag it for review or block it. If you run a public chatbot, this helps keep abusive or inappropriate messages from reaching your users or your model. Because the guardrails are built in, you do not need to wire up a separate moderation tool. Pricing is usage-based, so you pay as your app gets used.","useCases":["Screen chatbot inputs and outputs","Block or flag unsafe replies","Moderate user messages automatically"],"pricing":"Usage-based.","status":"ga","url":"https://www.digitalocean.com/products","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-agent-data","provider":"digitalocean","name":"Agent Data Services","category":"vector","tagline":"One place for your agents to store what they know, searchable several ways.","description":"Agent Data Services is a single store where AI agents keep the information they need to do their job. You can search it with SQL (the standard database question language), with vector search (finding things by meaning rather than exact words), or through an API. If you want your agent to remember past conversations or customer details, this is the kind of memory it would use. Having one store that you can reach several ways means you do not need a separate database for each style of search. It is in private preview, and pricing is not public.","useCases":["Give agents long-term memory","Store context agents can look up","Share one context store across agents"],"pricing":"Not public.","status":"preview","url":"https://www.digitalocean.com/products","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-paperspace","provider":"digitalocean","name":"Paperspace","category":"ai-platform","tagline":"Rent GPU computers and notebooks for AI work, from DigitalOcean.","description":"Paperspace is a cloud platform for machine learning, with machines that have GPUs (the chips that make AI training fast) and notebooks where you write and run code in small steps. DigitalOcean acquired it in 2023. If you want to fine-tune a model on your own data or try out ideas in a notebook, you can rent a GPU machine by the hour instead of buying hardware. It also includes a Kubernetes service, a tool for running many pieces of an app across machines. Pricing is hourly plus a subscription.","useCases":["Run GPU notebooks for model experiments","Fine-tune models on GPU machines","Rent GPU machines by the hour"],"pricing":"Hourly plus subscription.","status":"ga","url":"https://www.paperspace.com/","source":"https://www.digitalocean.com/products"},{"id":"digitalocean-cloudways","provider":"digitalocean","name":"Cloudways","category":"hosting","tagline":"Hosting that looks after WordPress and PHP sites for you.","description":"Cloudways is a managed hosting service, meaning it handles server setup and upkeep so you can focus on your site. It is a DigitalOcean-owned brand and runs on top of cloud providers. If you run a WordPress site for your business or an app written in PHP (a common web programming language), it gives you good hosting without learning server administration. Pick it over renting a bare server when you do not want to maintain one yourself. Plans start from $11 a month.","useCases":["Host WordPress sites without managing servers","Run PHP apps on managed servers","Launch a WordPress site quickly"],"pricing":"From $11/month.","status":"ga","url":"https://www.cloudways.com/","source":"https://www.digitalocean.com/products"},{"id":"hetzner-cloud-servers","provider":"hetzner","name":"Hetzner Cloud Servers","category":"vms","tagline":"Low-cost virtual servers in Europe, the US and Singapore.","description":"Hetzner Cloud Servers are virtual servers you rent by the hour, with a cap on the monthly price. Plans come as cost-optimized, regular performance and general purpose, which has a dedicated vCPU, meaning processor power reserved just for you. If you need a place to host a website or a dev server for testing, small plans start from about 4 EUR a month. Pick it over bigger clouds when low cost matters most. Note that the cost-optimized line was shown as unavailable on the cloud page when checked.","useCases":["Run websites on low-cost cloud servers","Set up test servers for development","Host apps in Europe, US or Singapore"],"pricing":"Hourly, capped monthly. Small plans from about 4 EUR/month.","status":"ga","url":"https://www.hetzner.com/cloud/","source":"https://www.hetzner.com/"},{"id":"hetzner-dedicated-servers","provider":"hetzner","name":"Dedicated Servers","category":"vms","also":["gpu"],"tagline":"Rent a whole physical server just for you, from about 40 EUR a month.","description":"Dedicated Servers are real physical computers that you rent by the month, so no one else shares the hardware. Hetzner offers them in several lines (AX, EX, SX, DX and GPU) in Germany and Finland, plus the US. If your app does heavy work, such as crunching lots of data, or you want to run your own cluster of servers, this gives you full control over the machine. Pick it over a virtual server when you want all of a machine's capacity to yourself. Prices start from about 40 EUR a month.","useCases":["Rent whole physical servers each month","Run heavy workloads on dedicated hardware","Build your own clusters on dedicated machines"],"pricing":"From about 40 EUR/month.","status":"ga","url":"https://www.hetzner.com/dedicated-rootserver/","source":"https://www.hetzner.com/"},{"id":"hetzner-server-auction","provider":"hetzner","name":"Server Auction","category":"vms","tagline":"Used dedicated servers at lower prices, with no setup fee.","description":"Server Auction sells refurbished dedicated servers, which are whole physical computers that have been used before, at reduced prices. There is no setup fee. If you want your own hardware for a side project or a database but a new server costs more than you want to spend, this is a cheaper way in. Prices start from about 30 EUR a month. Pick it over Hetzner's regular dedicated servers when saving money matters more than having brand new hardware.","useCases":["Buy refurbished dedicated servers cheaply","Get a dedicated server with no setup fee","Cut costs with used dedicated hardware"],"pricing":"From about 30 EUR/month.","status":"ga","url":"https://www.hetzner.com/sb/","source":"https://www.hetzner.com/"},{"id":"hetzner-object-storage","provider":"hetzner","name":"Object Storage","category":"object-storage","tagline":"Store files and backups online, with traffic bundled into the price.","description":"Object Storage is a place to keep files such as images, backups and website assets, organized in containers called buckets. It is S3-compatible, which means tools built for Amazon's S3 storage work with it. If your app lets people upload photos, or you need somewhere to keep database backups, you can put them here. Pricing is per terabyte, at about 6 EUR a month, with traffic bundled in. Pick it when you want S3-style storage with a simple per-terabyte price.","useCases":["Store backups in S3-compatible buckets","Host static files like images and downloads","Pay per terabyte with traffic included"],"pricing":"About 6 EUR/TB-month.","status":"ga","url":"https://www.hetzner.com/storage/object-storage/","source":"https://www.hetzner.com/"},{"id":"hetzner-storage-box","provider":"hetzner","name":"Storage Box","category":"block-storage","tagline":"Cheap remote storage for backups, reachable with common file tools.","description":"Storage Box is online storage built mainly for backups, in sizes up to 20 TB. You can reach it with common methods such as SFTP (secure file transfer), SMB, rsync (a file-copying tool), BorgBackup (a backup tool) and WebDAV, so many backup programs work with it directly. If you want an offsite copy of your app's data in case your main server fails, you can send nightly backups here. Pick it when your tools already speak these methods or you just want a simple place to copy files to. Plans start from about 4 EUR a month.","useCases":["Send offsite backups over SFTP or rsync","Store up to 20 TB of backups","Connect with WebDAV, SMB or BorgBackup"],"pricing":"From about 4 EUR/month.","status":"ga","url":"https://www.hetzner.com/storage/storage-box/","source":"https://www.hetzner.com/"},{"id":"hetzner-storage-share","provider":"hetzner","name":"Storage Share","category":"block-storage","also":["end-user"],"tagline":"Your own team file-sharing space, running Nextcloud.","description":"Storage Share is managed Nextcloud, a file-sharing tool that works like a private alternative to Dropbox or Google Drive. Hetzner runs it for you, and it is aimed at GDPR (European data protection law) needs. If your team needs one place to share documents with colleagues or clients, you can use it without setting up a server yourself. Pick it over plain storage when you want a ready-made file-sharing interface rather than raw space to fill. Plans start from about 5 EUR a month.","useCases":["Share files with your team","Run managed Nextcloud file sharing","Keep team files on GDPR-oriented storage"],"pricing":"From about 5 EUR/month.","status":"ga","url":"https://www.hetzner.com/storage/storage-share/","source":"https://www.hetzner.com/"},{"id":"hetzner-load-balancers","provider":"hetzner","name":"Cloud Load Balancers","category":"load-balancing","tagline":"Spread visitors across several servers so your site stays fast and online.","description":"Cloud Load Balancers sit in front of your servers and share incoming visitors between them. If your app gets a traffic spike after a launch, you add more servers and the load balancer divides the work. It checks that each server is healthy, so people are not sent to one that is down. It works at layer 4 and layer 7, meaning it can route plain network traffic or understand web requests, and it can handle HTTPS certificates too. Pricing starts from about 5 EUR/month.","useCases":["Spread traffic across several web servers","Add health checks to server pools","Serve HTTPS traffic with certificates"],"pricing":"From about 5 EUR/month.","status":"ga","url":"https://www.hetzner.com/cloud/load-balancer/","source":"https://www.hetzner.com/"},{"id":"hetzner-volumes","provider":"hetzner","name":"Cloud Volumes","category":"block-storage","tagline":"Extra disk space you can attach to your server and grow later.","description":"Cloud Volumes are extra storage drives that you attach to a Hetzner cloud server, a bit like plugging in an external hard drive. If your app's database or uploaded files are filling up your server's disk, you can add a volume instead of moving to a bigger server. They are fast SSD drives (the modern kind of disk) that connect over the network, and you can resize them as you grow. Pricing is about 0.05 EUR per GB per month.","useCases":["Add extra disk space to servers","Resize storage as your data grows","Attach network SSD volumes to servers"],"pricing":"About 0.05 EUR/GB-month.","status":"ga","url":"https://docs.hetzner.com/cloud/volumes/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-networks","provider":"hetzner","name":"Cloud Networks","category":"networking","tagline":"A private line between your servers, closed off from the public internet.","description":"Cloud Networks let your Hetzner servers talk to each other over a private connection instead of the open internet. If your app has a web server and a separate database server, you can keep the database reachable only from the web server. You set up private subnets, which are smaller address ranges inside your network, and routes that decide where traffic goes. It is free, so there is little reason to skip it once you run more than one server.","useCases":["Connect servers on a private network","Keep internal traffic off the public internet","Set routes between private subnets"],"pricing":"Free.","status":"ga","url":"https://docs.hetzner.com/cloud/networks/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-firewalls","provider":"hetzner","name":"Cloud Firewalls","category":"networking","tagline":"Choose which doors are open on your servers and block the rest.","description":"Cloud Firewalls decide what traffic is allowed to reach your servers and what gets blocked. If you run a website, you might open only the web ports (the numbered doors that services listen on) and close everything else, so strangers cannot poke at your database or admin tools. Stateful means the firewall remembers connections, so replies to requests your server started are let back in. You apply rule sets by label, so tagging a new server is enough to protect it. It is free.","useCases":["Block unwanted ports on servers","Apply firewall rules by server label","Lock down servers with stateful rules"],"pricing":"Free.","status":"ga","url":"https://docs.hetzner.com/cloud/firewalls/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-floating-ips","provider":"hetzner","name":"Floating and Primary IPs","category":"networking","tagline":"Fixed IP addresses you can move between servers when one fails.","description":"Floating and Primary IPs are public internet addresses, in both IPv4 and IPv6 formats, that you can reassign. If your app's server breaks or you want to swap in a new one, you point the same address at the replacement and your address stays the same. Floating IPs are the ones made to move between servers for failover, which means switching to a backup. That is quicker than changing DNS, the internet's address book, and waiting for the change to spread. IPv4 starts from about 0.50 EUR/month.","useCases":["Move an IP address to a backup server","Keep an app up during failover","Assign reusable IPv4 and IPv6 addresses"],"pricing":"IPv4 from about 0.50 EUR/month.","status":"ga","url":"https://docs.hetzner.com/cloud/floating-ips/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-backups-snapshots","provider":"hetzner","name":"Backups and Snapshots","category":"block-storage","tagline":"Save copies of your server so you can restore it if something goes wrong.","description":"Backups and Snapshots save a copy of your whole server so you can bring it back later. If an update breaks your app or you delete something by mistake, you restore the copy and carry on. Backups run automatically and add 20 percent to your server's price. Snapshots are copies you take yourself whenever you like, and you pay for the storage they use. Pick backups for a steady safety net, and snapshots for a quick save before a risky change.","useCases":["Restore servers after a failure","Take snapshots before risky updates","Add server backups for recovery"],"pricing":"Backups +20% of server cost.","status":"ga","url":"https://docs.hetzner.com/cloud/servers/backups-snapshots/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-dns","provider":"hetzner","name":"Hetzner DNS","category":"dns","tagline":"Free hosting for your domain's address book, managed in one place.","description":"Hetzner DNS is a free service that tells the internet where your domain points. DNS is the address book that turns a name like yourapp.com into the server that runs it. If you launch a site on a Hetzner server, you can manage your domain's records in the same Cloud console, or through the API, which lets code make changes for you. It costs nothing, so it is a simple pick when your servers already live at Hetzner.","useCases":["Manage domain records in one place","Point domains to your servers","Control DNS through the console or API"],"pricing":"Free.","status":"ga","url":"https://docs.hetzner.com/networking/dns/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-managed-certificates","provider":"hetzner","name":"SSL Certificates","category":"certificates","tagline":"Free padlock for your site: HTTPS certificates for your load balancers.","description":"SSL Certificates give your site the padlock and the https address that visitors expect. If you put a load balancer in front of your app, you can have Hetzner create a certificate through Let's Encrypt, a free certificate provider, right from the Cloud console. You can also upload one you already have. The encryption is handled at the load balancer, so you do not have to set it up on each server. It is free.","useCases":["Add free SSL certificates to load balancers","Upload your own certificates for HTTPS","Secure sites with Let's Encrypt certificates"],"pricing":"Free.","status":"ga","url":"https://docs.hetzner.com/networking/certificates/overview/","source":"https://www.hetzner.com/"},{"id":"hetzner-managed-server","provider":"hetzner","name":"Managed Server","category":"vms","tagline":"A server that Hetzner looks after for you, with support included.","description":"Managed Server is hosting where Hetzner's team looks after the server for you, so you do not handle the technical upkeep yourself. If you want your site online but nobody on your team wants to deal with updates and server settings, this hands that work to the provider. Admin is included, which sets it apart from a regular cloud server where you do everything yourself. It suits anyone who wants hands-off hosting. For pricing, contact Hetzner or see their site.","useCases":["Hand off server admin to Hetzner","Host apps without managing the server","Get support with fully managed hosting"],"pricing":"Contact or see site.","status":"ga","url":"https://www.hetzner.com/managed-server/","source":"https://www.hetzner.com/"},{"id":"hetzner-webhosting","provider":"hetzner","name":"Webhosting","category":"hosting","tagline":"Simple, low-cost website hosting for small sites, without running a server.","description":"Webhosting is classic shared hosting, where your site lives on a server that many other sites share. If you are putting up a small business site, a blog or a simple page, you upload your files and you are live, with no server to set up. It supports PHP, a common language for websites, and MySQL, a popular database. Pick it over a cloud server when your site is small and you want the simplest, cheapest option. Plans start from about 2 EUR/month.","useCases":["Host small PHP and MySQL sites","Launch a basic website on shared hosting","Run classic PHP apps cheaply"],"pricing":"From about 2 EUR/month.","status":"ga","url":"https://www.hetzner.com/webhosting/","source":"https://www.hetzner.com/"},{"id":"hetzner-domain-registration","provider":"hetzner","name":"Domain Registration","category":"domains","tagline":"Buy your domain name from Hetzner and manage its settings in one place.","description":"Domain Registration lets you buy a web address, like yourapp.com, from Hetzner. If you are launching a product and need a name, you can register it and manage its DNS, the settings that point the name at your server, from the same account. That keeps your domain, servers and records together instead of spread across several companies. The price depends on the TLD, which is the ending of the domain such as .com or .org, so check the cost for the one you want.","useCases":["Register new domain names","Buy domains and manage DNS together","Get a domain for your new app"],"pricing":"Per TLD.","status":"ga","url":"https://www.hetzner.com/registrationrobot/","source":"https://www.hetzner.com/"},{"id":"vultr-cloud-compute","provider":"vultr","name":"Cloud Compute","category":"vms","tagline":"Low-cost virtual servers you can start in many places around the world.","description":"Cloud Compute gives you virtual machines, which are slices of a bigger computer that act like your own private server. If you are launching a web app or want a place to build and test, you can rent one and run your code there. These are shared-CPU machines, meaning you share the processor with other customers, which keeps the price down. They are general purpose and available in 32 regions, so you can place your app close to your users. Prices start from $2.50/month.","useCases":["Run general web apps on shared VMs","Launch dev environments in 32 regions","Start small virtual machines from $2.50"],"pricing":"From $2.50/month.","status":"ga","url":"https://www.vultr.com/products/cloud-compute/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-optimized-cloud-compute","provider":"vultr","name":"Optimized Cloud Compute","category":"vms","tagline":"Rented servers with reserved processing power for heavy database work.","description":"Optimized Cloud Compute gives you virtual servers (VMs, which are slices of a bigger machine) with dedicated processor cores, so you are not sharing that capacity with other customers. Plans are tuned for processing power, memory or storage. If your app has a busy database, or you run analytics or compile code regularly, this is a good place to look. Pick it over a basic shared server when slowdowns from other users would hurt. Plans start at $28 a month.","useCases":["Run busy databases on dedicated CPUs","Compile code faster on dedicated vCPUs","Power analytics with optimized machines"],"pricing":"From $28/month.","status":"ga","url":"https://www.vultr.com/products/optimized-cloud-compute/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-vx1","provider":"vultr","name":"VX1 Cloud Compute","category":"vms","tagline":"Fast rented servers with built-in speedy storage for data-heavy work.","description":"VX1 Cloud Compute is a line of virtual servers (VMs) with dedicated processor cores, built on AMD EPYC chips and fitted with local NVMe, a very fast type of storage attached right to the machine. It is tuned for work that moves a lot of data in and out of storage and across the network. If you are building a distributed system, meaning an app spread across many machines working together, this fits well. Pick it over a general-purpose server when storage speed and bandwidth are your bottleneck. Check Vultr's pricing page for current rates.","useCases":["Run storage-heavy workloads on fast NVMe","Handle high bandwidth traffic","Run distributed systems on AMD servers"],"pricing":"See pricing.","status":"ga","url":"https://www.vultr.com/products/vx1-cloud-compute/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-cloud-gpu","provider":"vultr","name":"Cloud GPU","category":"gpu","tagline":"Rent AI-ready graphics chips by the hour to train and run models.","description":"Cloud GPU lets you rent graphics chips (GPUs), the hardware that handles the heavy math behind modern AI. You can get NVIDIA and AMD chips as virtual servers, whole physical machines, or clusters of many. If you are training your own AI model, or running one to answer requests from your users (called inference), this is the kind of hardware you need. Pick it when you want to choose your own chips and setup. It includes AMD Instinct MI355X GPUs from $2.29 per GPU per hour, billed hourly.","useCases":["Train AI models on GPUs","Run AI inference on GPU machines","Rent GPUs hourly as VMs or clusters"],"pricing":"Hourly per GPU.","status":"ga","url":"https://www.vultr.com/products/cloud-gpu/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-bare-metal","provider":"vultr","name":"Bare Metal","category":"vms","tagline":"A whole physical server to yourself, ready on demand.","description":"Bare Metal gives you an entire physical server that only you use, set up automatically when you order it. There is no virtualization layer, which means no one else shares the machine. If you are building something where every millisecond counts, like a real-time trading tool or a multiplayer game backend, that consistency matters. Pick it over a regular virtual server when you want full control of the hardware and steady, predictable speed. Plans start at $120 a month.","useCases":["Run latency-sensitive systems on physical servers","Deploy automated single-tenant servers on demand","Rent physical servers on demand"],"pricing":"From $120/month.","status":"ga","url":"https://www.vultr.com/products/bare-metal/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-kubernetes","provider":"vultr","name":"Vultr Kubernetes Engine","category":"containers","tagline":"Run container apps on managed Kubernetes, with no fee for the control layer.","description":"Vultr Kubernetes Engine is a managed version of Kubernetes, a popular system for running and scaling apps packaged as containers (self-contained bundles of your code and what it needs). If your product is already split into containers, this handles the coordination work so you do not have to build it yourself. The control plane, the part that directs everything, is free, and you pay only for the nodes (the servers that run your apps). It also supports GPU node pools, handy if some of your workloads need graphics chips for AI.","useCases":["Run container apps on managed Kubernetes","Add GPU node pools for AI workloads","Use a free Kubernetes control plane"],"pricing":"Pay for nodes.","status":"ga","url":"https://www.vultr.com/kubernetes/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-serverless-inference","provider":"vultr","name":"Serverless Inference","category":"ai-models","also":["vector"],"tagline":"Call AI models over the web, with built-in storage to chat with your data.","description":"Serverless Inference lets you use AI models through an API, a simple web connection, without setting up or managing any servers. It offers a catalog of language models (the chatbot kind), vision models and embedding models. If you want a chat feature in your app that answers from your own documents, its built-in vector storage supports RAG, a method where the AI looks up your content before it replies. Pick it when you want AI features quickly without running your own hardware. Pricing is based on usage.","useCases":["Add chat features with hosted AI models","Build search features with vector storage","Pay per use for embedding and vision models"],"pricing":"Usage-based.","status":"ga","url":"https://www.vultr.com/products/serverless-inference/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-block-storage","provider":"vultr","name":"Block Storage","category":"block-storage","tagline":"Extra disk space you attach to your servers and resize as you grow.","description":"Block Storage is like adding an extra hard drive to your cloud server. You can choose NVMe volumes, which are very fast, or HDD volumes, which are slower and cheaper. If you run your own database, this is where its data would live, so it stays safe even if you rebuild the server. Volumes can be resized, and snapshots (saved copies of a volume at a moment in time) let you go back if something breaks. Pricing starts at $0.10 per GB per month.","useCases":["Attach database disks to servers","Take snapshots of volumes","Resize NVMe or HDD volumes as needed"],"pricing":"From $0.10/GB-month.","status":"ga","url":"https://www.vultr.com/products/block-storage/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-object-storage","provider":"vultr","name":"Object Storage","category":"object-storage","tagline":"Store files, images and backups online, with Amazon S3 compatibility.","description":"Object Storage is a place to keep large numbers of files, such as photos, videos and backups, and reach them over the web. It is S3-compatible, so it works with tools built for Amazon S3, a widely used storage service. If your app lets people upload images or you need cheap, safe backups, this is a natural fit. Files are private by default. Pricing is tiered, and there is an archival tier for data you rarely touch. Plans start at $6 a month.","useCases":["Store user uploads and media files","Keep backups of your data","Archive old files at lower cost"],"pricing":"From $6/month.","status":"ga","url":"https://www.vultr.com/products/object-storage/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-file-system","provider":"vultr","name":"File System","category":"block-storage","tagline":"A shared folder that many of your servers can use at the same time.","description":"File System is shared storage that works like a network folder. It uses NFS, a standard way for machines to share files over a network, so several of your servers can mount it and see the same files. If you run more than one server and they all need the same uploads, assets or config files, this keeps everything in one place. Pick it over Block Storage when you need multiple machines reading and writing the same data. Pricing is based on usage.","useCases":["Share one folder between servers","Mount the same files on several servers","Keep team files in one place"],"pricing":"Usage-based.","status":"ga","url":"https://docs.vultr.com/products","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-storage-gateway","provider":"vultr","name":"Storage Gateway","category":"block-storage","tagline":"A bridge that lets Bare Metal servers use Vultr's shared file storage.","description":"Storage Gateway connects your Bare Metal servers to Vultr File System using NFS, a standard way of sharing files over a network. If you run physical servers and want them to read and write the same shared files, this is the link between the two. It saves you from building that connection yourself. This service is in preview, and it is free during the API pilot program, so details may change before it becomes generally available.","useCases":["Share files with bare metal servers","Connect bare metal machines to shared storage","Test shared storage at no cost during pilot"],"pricing":"Free in pilot.","status":"preview","url":"https://docs.vultr.com/product-overviews","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-postgres","provider":"vultr","name":"Managed PostgreSQL","category":"sql","tagline":"A ready-to-run Postgres database that Vultr looks after for you.","description":"Managed PostgreSQL gives you Postgres, a very popular database, without the chore of running it yourself. If you are building an app that stores users, orders or content, this can be its main database. It includes PgBouncer, which manages connections so lots of visitors do not overwhelm the database, plus replicas (extra copies) for reading data. Automated backups and HA (high availability, meaning it keeps running if one machine fails) are included. Plans start at $15 a month.","useCases":["Run a PostgreSQL database for your app","Add read replicas for busy traffic","Get automatic backups and failover"],"pricing":"From $15/month.","status":"ga","url":"https://www.vultr.com/products/managed-databases/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-mysql","provider":"vultr","name":"Managed MySQL","category":"sql","tagline":"A ready-to-run MySQL database that Vultr looks after for you.","description":"Managed MySQL gives you MySQL, one of the most widely used databases, without having to set it up or maintain it. If you are running a website or web app that stores things like customers, posts or products, this can hold that data. Replicas (extra copies of your database) can help with handling more reads, and automated backups protect you if something goes wrong. Pick it when your app or tools, such as many website builders, already expect MySQL. Plans start at $15 a month.","useCases":["Run a MySQL database for a website","Store web data in a managed database","Add read replicas for more traffic"],"pricing":"From $15/month.","status":"ga","url":"https://www.vultr.com/products/managed-databases/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-valkey","provider":"vultr","name":"Managed Valkey","category":"kv-cache","tagline":"A fast memory store for caching, compatible with Redis, run for you.","description":"Managed Valkey is a fast temporary data store, compatible with Redis (a popular caching tool), that Vultr runs for you. If your app feels slow because it keeps asking the database for the same things, like a product list or a logged-in user's details, you can keep those answers in Valkey and serve them instantly. It supports ACLs (rules for who can do what) and read replicas, which are extra copies that share the load of lookups. Pick it when you want Redis-style speed without installing or updating anything yourself. Plans start at $15/month.","useCases":["Cache frequently used data for speed","Set access rules for each user","Add read replicas for heavy reads"],"pricing":"From $15/month.","status":"ga","url":"https://www.vultr.com/products/managed-databases/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-kafka","provider":"vultr","name":"Managed Apache Kafka","category":"queues","tagline":"A managed stream of live events between your apps, with Kafka run for you.","description":"Managed Apache Kafka moves a constant flow of events, like clicks, orders or sensor readings, from one part of your system to another. Vultr runs the clusters for you, and it includes Connect (ready-made links to other tools) and Schema Registry (a place that keeps your data formats consistent). If your product needs to feed live activity into analytics or several other services, this is the pipeline you would build. Clusters are available in 33+ locations. It starts at about $100/month, so it suits real streaming needs.","useCases":["Stream events from one app to another","Move data through live pipelines","Keep data formats consistent across services"],"pricing":"From about $100/month.","status":"ga","url":"https://www.vultr.com/products/managed-databases/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-load-balancer","provider":"vultr","name":"Load Balancers","category":"load-balancing","tagline":"Spread visitors across several servers so your app stays up.","description":"Load Balancers sit in front of your servers and share incoming traffic between them. If your app gets busy after a launch or a mention in the press, you can add more servers and let the load balancer send each visitor to one that is ready. They work at Layer 4 and Layer 7 (basic traffic level or web-request level) and run health checks, so a server that stops responding gets skipped. Choose one when a single server is no longer enough. They start at $10/month.","useCases":["Spread traffic across several servers","Check server health before routing traffic","Add servers as demand grows"],"pricing":"From $10/month.","status":"ga","url":"https://www.vultr.com/products/load-balancer/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-cdn","provider":"vultr","name":"Vultr CDN","category":"cdn","tagline":"Serve images and files from locations close to your visitors.","description":"Vultr CDN keeps copies of your static files, like images, scripts and stylesheets, at edge locations (servers spread around the world) so they load quickly wherever your visitors are. If your site has lots of photos or downloads, a CDN means your main server does less work and pages appear faster. It supports pull mode, where it fetches files from your site on its own, and push mode, where you upload files to it directly. Pick it when you already run on Vultr and want fast file delivery in the same account. Pricing is usage-based.","useCases":["Serve images and scripts from nearby locations","Cache static files at the edge","Speed up downloads for far-away visitors"],"pricing":"Usage-based.","status":"ga","url":"https://www.vultr.com/products/cdn/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-dns","provider":"vultr","name":"Vultr DNS","category":"dns","tagline":"Free hosting for your domain's address book, so your name finds your app.","description":"Vultr DNS is the address book for your domain name, telling browsers which server to visit when someone types your web address. If you just bought a domain for your new product, you create a zone (the set of records for that domain) here and point it at your app. It is managed for you, so there is nothing to install or keep running. Pick it when you want your domain settings in the same place as your servers, especially since it costs nothing.","useCases":["Point your domain to your servers","Manage DNS records for your domain","Host domain records at no cost"],"pricing":"Free.","status":"ga","url":"https://www.vultr.com/products/dns/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-networking","provider":"vultr","name":"VPC, Firewall and Reserved IPs","category":"networking","tagline":"Private networks, firewall rules and fixed IP addresses for your servers.","description":"This bundle covers the basic plumbing for your servers: VPC (a private network that only your machines can use), firewall groups (rules about what traffic is allowed in) and reserved IPs (addresses that stay yours even if you swap servers). If your database should only talk to your app and never the open internet, you would put both on a private subnet and lock everything else down. VPC 2.0 and firewalls are free, and the set is mostly free overall. Pick it as the default way to keep a Vultr setup tidy and safe.","useCases":["Isolate servers on a private network","Control traffic with firewall groups","Keep a fixed IP address for a server"],"pricing":"Mostly free.","status":"ga","url":"https://www.vultr.com/features/","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-backups","provider":"vultr","name":"Automatic Backups","category":"block-storage","tagline":"Automatic scheduled copies of your server so you can recover from mistakes.","description":"Automatic Backups take scheduled copies of your Vultr server so you can roll back if something breaks. If an update goes wrong or someone deletes the wrong folder, you restore the saved copy instead of rebuilding from scratch. The price is simple: 20 percent of what the server itself costs, added on top. Choose it when you want a safety net that runs on its own and you do not want to write your own backup scripts.","useCases":["Schedule automatic backups of instances","Recover a server after something breaks","Add backup protection for each server"],"pricing":"+20% of instance price.","status":"ga","url":"https://docs.vultr.com/products","source":"https://docs.vultr.com/product-overviews"},{"id":"vultr-container-registry","provider":"vultr","name":"Container Registry","category":"containers","tagline":"A private place to store the packaged versions of your app.","description":"Container Registry is a private shelf for container images, which are packaged copies of your app that are ready to run anywhere. If you build your app into a container and deploy it several times a week, you push each version here and your servers pull it when needed. It follows the OCI standard (a common format for these packages) and serves Vultr Kubernetes, a service that runs many containers for you. Pick it when you already deploy on Vultr and want your images kept close to where they run. It starts at $5/month.","useCases":["Store your private container images","Hold images for Kubernetes deployments","Keep app builds ready to deploy"],"pricing":"From $5/month.","status":"ga","url":"https://www.vultr.com/products/container-registry/","source":"https://docs.vultr.com/product-overviews"},{"id":"flyio-machines","provider":"flyio","name":"Fly Machines","category":"vms","also":["containers"],"tagline":"Small, quick-starting servers that run your app in 18+ regions worldwide.","description":"Fly Machines are small virtual servers that run your app packaged as a container (a ready-to-run bundle of your code), and they boot fast. If your API serves people in Europe, Asia and the US, you can run copies in 18+ regions so each person talks to one nearby, or use them for background jobs like sending emails. They are billed per second while running, and you can stop or suspend them when idle. Each is isolated at the hardware level. They are Fly's core compute product, from $2.19/month for the shared-1x 256MB size.","useCases":["Run APIs close to your users","Process background jobs on demand","Pay only while machines run"],"pricing":"From $2.19/month for shared-1x 256MB.","status":"ga","url":"https://fly.io/machines.md","source":"https://fly.io/pricing"},{"id":"flyio-sprites","provider":"flyio","name":"Sprites","category":"agents","tagline":"Safe, saved computers where AI agents can run code and pick up later.","description":"Sprites are persistent sandboxes, meaning separate computers that keep their files and state, where an AI agent can work without touching anything else. If your product has a coding assistant that writes and runs code for users, or you need to run code you do not trust, a Sprite keeps it contained, and checkpoints let you save and return to a known state. They sleep when idle and bill only for active CPU, memory and hot storage, so a quiet one costs little. Pricing is $0.0385 per CPU-hour and $0.021875 per GB-hour.","useCases":["Run coding agents in isolated computers","Execute untrusted code safely","Pause idle sandboxes and save checkpoints"],"pricing":"$0.0385/CPU-hour, $0.021875/GB-hour.","status":"ga","url":"https://fly.io/sprites.md","source":"https://fly.io/pricing"},{"id":"flyio-managed-postgres","provider":"flyio","name":"Managed Postgres (MPG)","category":"sql","tagline":"A production-ready Postgres database that Fly runs, backs up and protects.","description":"Managed Postgres is Fly's own service for running Postgres, a popular database, so you do not have to look after it yourself. If your app needs to store users, orders or anything that must not disappear, it gives you HA (high availability, meaning a standby takes over if something fails), backups and pooling (sharing connections so many requests do not overwhelm the database). It replaces the older self-managed Fly Postgres, so it is the one to pick for a new Fly app. Plans run from Basic at $38/month up to Performance at $1,922/month, plus $0.28/GB.","useCases":["Run production Postgres with backups","Scale database plans as the app grows","Use built-in connection pooling"],"pricing":"$38/month (Basic) to $1,922/month plus $0.28/GB.","status":"ga","url":"https://fly.io/mpg.md","source":"https://fly.io/pricing"},{"id":"flyio-volumes","provider":"flyio","name":"Fly Volumes","category":"block-storage","tagline":"Fast local disk space that keeps your data when a Machine restarts.","description":"Fly Volumes are persistent disks, meaning storage that keeps your files, that you attach to a Fly Machine. If you run your own database or an app that saves files, a volume stops that data from vanishing when the Machine restarts. They use NVMe, a very fast type of drive, and sit locally on the same hardware. One volume attaches to a single Machine, so it is not shared between several. Pick it when you run your own database and need quick storage, at $0.15/GB-month.","useCases":["Attach persistent disks to a Machine","Run your own database on a disk","Store data that survives restarts"],"pricing":"$0.15/GB-month.","status":"ga","url":"https://docs.fly.io/volumes/overview/","source":"https://fly.io/pricing"},{"id":"flyio-fks","provider":"flyio","name":"Fly Kubernetes (FKS)","category":"containers","tagline":"Kubernetes on Fly, with the management layer handled for you.","description":"Fly Kubernetes (FKS) is Fly's managed version of Kubernetes, a popular system for organizing and running many app containers. Your pods (the running pieces of your app) run as Fly Machines, Fly's small servers. If your team already has Kubernetes setup files and wants to run them on Fly, this lets you keep that workflow instead of starting over. Pick it when you specifically need Kubernetes tools. It costs $75 per month per cluster, and compute and volumes (the disk storage) are billed separately.","useCases":["Run Kubernetes workflows on Fly","Deploy pods as Fly Machines","Manage a cluster for container apps"],"pricing":"$75/month per cluster.","status":"ga","url":"https://fly.io/docs/kubernetes/","source":"https://fly.io/pricing"},{"id":"flyio-tigris","provider":"flyio","name":"Tigris Object Storage (partner)","category":"object-storage","tagline":"Store files and images worldwide, billed through your Fly invoice.","description":"Tigris Object Storage is a place to keep files like images, videos and AI model files. It works with the S3 format (Amazon's standard way of talking to file storage), so many existing tools can connect to it. If your app lets people upload photos, or you need somewhere for large model weights (the files an AI model runs from), this fits. It spreads your files globally. It is a partner service run by Tigris Data, not Fly, and you pay Tigris's list prices on your Fly invoice.","useCases":["Store user uploads and large files","Keep large model weights in storage","Store files on a global network"],"pricing":"Tigris list pricing.","status":"ga","url":"https://docs.fly.io/tigris/","source":"https://fly.io/pricing"},{"id":"flyio-upstash-redis","provider":"flyio","name":"Upstash Redis (partner)","category":"kv-cache","tagline":"Fast caching and rate limiting, run by Upstash for your Fly apps.","description":"Upstash Redis is a fast storage space, built on Redis (a popular tool that keeps data in memory), that you can add to your Fly apps. If your app keeps asking the database the same question, you can save the answer here so pages load faster. It is also handy for rate limits, which stop one visitor from hitting your app too many times in a short window. Pick it when you want Redis without running it yourself. It is a partner service operated by Upstash, at Upstash list pricing.","useCases":["Cache results for faster responses","Limit how often users make requests","Use Redis without running a server"],"pricing":"Upstash list pricing.","status":"ga","url":"https://docs.fly.io/upstash/redis/","source":"https://fly.io/pricing"},{"id":"flyio-networking","provider":"flyio","name":"Anycast, Fly Proxy and Private Networking","category":"networking","also":["load-balancing"],"tagline":"Built-in traffic routing, HTTPS and a private network for your apps.","description":"This is the networking built into Fly. It sends visitors to your app, handles TLS (the encryption behind the HTTPS padlock), and links your apps over a private network using WireGuard, a modern secure-tunnel technology. If your app runs in several regions, Anycast (one address that sends each visitor to a nearby copy) and Fly Proxy (Fly's load balancer, which spreads traffic across your servers) do the routing. Shared IPv4 and IPv6 addresses are included, and a dedicated IPv4 address costs $2 per month.","useCases":["Route visitors to the nearest server","Connect servers on a private network","Add a dedicated IP address for $2"],"pricing":"Dedicated IPv4 $2/month.","status":"ga","url":"https://fly.io/docs/networking/","source":"https://fly.io/pricing"},{"id":"render-web-services","provider":"render","name":"Web Services","category":"hosting","also":["containers"],"tagline":"Put your web app or API online straight from Git or Docker.","description":"Web Services is Render's way to run the backend of a website or an API (the part of your app that answers requests from your front end) on the public internet. If you have a full-stack app in a Git repository, you connect it and Render deploys it, or you can supply a Docker package (your app bundled with everything it needs). It includes HTTP/2 and TLS, which keep connections fast and secure, plus autoscaling, which adds capacity when traffic rises, on Pro and above. There is a free tier, and paid plans start at $7 per month.","useCases":["Host APIs and full-stack web apps","Deploy straight from Git repositories","Scale instances automatically with demand"],"pricing":"Free tier; from $7/month.","status":"ga","url":"https://render.com/docs/web-services","source":"https://render.com/pricing"},{"id":"render-static-sites","provider":"render","name":"Static Sites","category":"hosting","also":["cdn"],"tagline":"Free hosting for simple websites, served from servers worldwide.","description":"Static Sites hosts websites made of ready-made files, like a marketing site or a front end with no server code behind it. If you are launching a landing page, you can publish it on a CDN (a network of servers around the world that delivers pages quickly). It is free and includes preview environments, which give each change its own test link before it goes live. Pick it over a Web Service when your site does not need to run code on a server.","useCases":["Host marketing sites on a global network","Preview changes before going live","Publish frontend sites at no cost"],"pricing":"Free.","status":"ga","url":"https://render.com/docs/static-sites","source":"https://render.com/pricing"},{"id":"render-private-services","provider":"render","name":"Private Services","category":"containers","tagline":"Run internal services that only your own apps can reach.","description":"Private Services are programs that run on Render but are not open to the public internet. They work like Web Services without public access, so only your other services on your private network can talk to them. If your app has an internal API, say one that handles billing logic for your main app, this keeps it hidden from outsiders. Pick it over a Web Service when the public should never reach it. Pricing starts at $7 per month.","useCases":["Run internal APIs off the public internet","Keep backend services hidden from visitors","Connect internal services over a private network"],"pricing":"From $7/month.","status":"ga","url":"https://render.com/docs/private-services","source":"https://render.com/pricing"},{"id":"render-background-workers","provider":"render","name":"Background Workers","category":"queues","tagline":"Always-running helpers that process jobs behind the scenes.","description":"Background Workers are programs that run all the time on Render without a web address. They suit long-running work, like picking jobs off a queue (a waiting line of tasks) and handling them one by one. If your app sends welcome emails or resizes uploaded images, a worker can do it so your visitors never wait. Pick it over a Cron Job when work arrives at unpredictable times instead of on a schedule. Plans start at $7 per month.","useCases":["Process jobs from a queue continuously","Run long tasks without a web port","Run always-on queue consumers"],"pricing":"From $7/month.","status":"ga","url":"https://render.com/docs/background-workers","source":"https://render.com/pricing"},{"id":"render-cron-jobs","provider":"render","name":"Cron Jobs","category":"workflows","tagline":"Run a command on a schedule, like every night, and pay only while it runs.","description":"Cron Jobs run a command on a schedule you set, such as every night at 2 a.m. You write the schedule as a cron expression (a short code for repeating times), and Render keeps logs of each run so you can check what happened. If your app needs to clear out old data or run a nightly task, this handles it. Pick it over a Background Worker when the task is occasional, since you pay for compute only while it runs.","useCases":["Run scheduled commands on a timetable","Run nightly tasks automatically","Check logs for each scheduled run"],"pricing":"Per-minute compute.","status":"ga","url":"https://render.com/docs/cronjobs","source":"https://render.com/pricing"},{"id":"render-workflows","provider":"render","name":"Render Workflows","category":"workflows","tagline":"Long-running tasks that can run up to 24 hours each. In beta.","description":"Render Workflows runs long jobs, with each task allowed to run for up to 24 hours. If your app starts an AI job, like working through a batch of documents, or crunches a big pile of data, this is where that work can live. Pick it when a job is far too long for a normal web request. Your plan sets how many tasks can run at once, and the compute each task uses is billed. It is currently in beta, so expect changes.","useCases":["Run long tasks up to 24 hours","Run AI jobs in the background","Process data in long-running jobs"],"pricing":"Plan-based concurrency; task compute billed.","status":"beta","url":"https://render.com/docs/workflows","source":"https://render.com/pricing"},{"id":"render-postgres","provider":"render","name":"Render Postgres","category":"sql","tagline":"A ready-to-use database for your app, with backups and failover.","description":"Render Postgres is a managed database, meaning Render runs it for you, built on Postgres, a very popular database. If your app needs to remember users, orders or posts, this is where that data lives. It supports point-in-time recovery (rewinding to an earlier moment after a mistake), replicas (extra copies of your data) and high availability (a standby that can take over if something fails). Plans range from a free 256MB of memory up to 1 TB, and paid plans start at $6 per month.","useCases":["Store app data in a database","Restore data to an earlier point","Add copies for high availability"],"pricing":"Free; paid from $6/month.","status":"ga","url":"https://render.com/docs/postgresql","source":"https://render.com/pricing"},{"id":"render-key-value","provider":"render","name":"Render Key Value","category":"kv-cache","tagline":"A fast memory store for caching and queues, compatible with Redis.","description":"Render Key Value is a managed store that keeps data in memory for very fast reads. It is compatible with Redis, a popular tool for this, so existing Redis code should work with it. If your app keeps asking the database the same question, you can save the answer here to speed things up. It also works as a queue, a waiting line for background jobs. Pick it when you want a cache on the same platform as the rest of your Render services. There is a free 25 MB plan, and paid plans start at $10 per month.","useCases":["Speed up apps with caching","Hold tasks in a queue","Store short-lived data quickly"],"pricing":"Free 25 MB; from $10/month.","status":"ga","url":"https://render.com/docs/key-value","source":"https://render.com/pricing"},{"id":"render-disks","provider":"render","name":"Persistent Disks","category":"block-storage","tagline":"Fast storage that keeps your files when your app restarts or redeploys.","description":"Persistent Disks are fast solid-state drives (SSDs) that you attach to your Render web services, background workers or private services, so saved files stick around when the app restarts or redeploys. If your app lets people upload photos or documents, a disk gives those files a home right next to your code. It also works if you want to run your own database instead of using a managed one. Pick it when you want simple file storage tied to one service. It costs $0.25 per GB each month.","useCases":["Save uploaded files on disk","Run your own databases","Attach storage to web services"],"pricing":"$0.25/GB-month.","status":"ga","url":"https://render.com/docs/disks","source":"https://render.com/pricing"},{"id":"render-preview-environments","provider":"render","name":"Preview Environments","category":"devtools","tagline":"A full test copy of your app and database for every proposed code change.","description":"Preview Environments spin up a complete copy of your app and its database for each pull request, which is a proposed change to your code. They are built from Blueprints, Render's files that describe how your setup fits together. If you are about to change how checkout works, you can try it out for real before it goes live. Pick this over testing on your live site or one shared test site, since every change gets its own space. You pay for the resources each copy uses.","useCases":["Test each pull request separately","Copy the app and database","Spin up full app copies per change"],"pricing":"Pay for resources.","status":"ga","url":"https://render.com/docs/preview-environments","source":"https://render.com/pricing"},{"id":"railway-services","provider":"railway","name":"Railway Services","category":"hosting","also":["containers"],"tagline":"Put your app online from GitHub, Docker or a template, billed per second.","description":"Railway Services is where your app actually runs. You connect GitHub, a Docker image (a packaged-up app) or a ready-made template, and Railway runs it for you. If you are launching a full-stack app or an API, this is where it lives. Billing is by usage, per second, at $20 per vCPU-month (processor power) and $10 per GB-month of RAM (working memory), so you pay for what your app uses. Plans are free, Hobby at $5, or Pro at $20 plus usage, and Pro includes unlimited seats for your team.","useCases":["Host full-stack apps","Deploy APIs from GitHub","Deploy apps from Docker images"],"pricing":"Free; Hobby $5; Pro $20 plus usage.","status":"ga","url":"https://railway.com/","source":"https://railway.com/pricing.md"},{"id":"railway-postgres","provider":"railway","name":"Postgres","category":"sql","tagline":"A ready-to-go Postgres database with backups, set up in one click.","description":"Postgres is a popular database for storing your app's data, like user accounts and orders, and on Railway you can add it in one click. It runs as a service inside your project, right next to your app. It includes backups and PITR (point-in-time recovery), which lets you roll the database back to an earlier moment if something goes wrong. Pick it when you want your app and its database managed in one place. Pricing is based on usage.","useCases":["Run an app database","Back up database automatically","Restore data to earlier points"],"pricing":"Usage-based.","status":"ga","url":"https://docs.railway.com/databases/postgresql","source":"https://railway.com/pricing.md"},{"id":"railway-mysql","provider":"railway","name":"MySQL","category":"sql","tagline":"A ready-to-go MySQL database, set up in one click.","description":"MySQL is a widely used database for storing your app's data, and on Railway you add it in one click from a template. It runs as a service in your project, alongside your app. If you are building a web app, or using software that expects MySQL, such as WordPress, this saves you from setting up a server yourself. Pick it over Postgres when your tools specifically ask for MySQL, otherwise either one will do the job. Billing is by usage.","useCases":["Run MySQL databases for web apps","Launch a database from a template","Store data for web apps"],"pricing":"Usage-based.","status":"ga","url":"https://docs.railway.com/databases/mysql","source":"https://railway.com/pricing.md"},{"id":"railway-redis","provider":"railway","name":"Redis","category":"kv-cache","tagline":"One-click Redis, a very fast place to keep data your app needs often.","description":"Redis is a very fast database that keeps data in memory, which makes it a common choice for a cache: a quick-access shelf for things your app asks for again and again. On Railway you add it in one click from a template, and it runs as a service in your project. If your pages feel slow because they keep looking up the same information, a cache can speed them up. Pick it as a companion to your main database, not a replacement for it. Billing is by usage.","useCases":["Cache data for faster responses","Launch Redis from a template","Speed up your app's cache"],"pricing":"Usage-based.","status":"ga","url":"https://docs.railway.com/databases/redis","source":"https://railway.com/pricing.md"},{"id":"railway-mongodb","provider":"railway","name":"MongoDB","category":"nosql","tagline":"One-click MongoDB, a database that stores data as flexible documents.","description":"MongoDB is a database that stores information as flexible documents, a bit like JSON files, instead of strict tables. On Railway you add it in one click from a template, and it runs as a service in your project. If your app's data varies from item to item, like user profiles with different fields, this style can feel natural. Pick it over Postgres or MySQL when your data does not fit neatly into rows and columns. Billing is by usage.","useCases":["Store document-style data","Launch MongoDB from a template","Keep flexible data records"],"pricing":"Usage-based.","status":"ga","url":"https://docs.railway.com/databases/mongodb","source":"https://railway.com/pricing.md"},{"id":"railway-volumes","provider":"railway","name":"Volumes","category":"block-storage","tagline":"Storage that keeps your service's files when it restarts.","description":"Volumes are disks you attach to your Railway services so files survive restarts and redeploys. If you run something that has to remember things, like a database you manage yourself or an app that saves files on its own drive, a volume gives it a permanent place to write. Pick it over Buckets when the files belong right next to one service instead of in separate storage. It costs $0.15 per GB each month.","useCases":["Keep files on persistent disks","Attach storage to services","Keep data that survives restarts"],"pricing":"$0.15/GB-month.","status":"ga","url":"https://docs.railway.com/volumes","source":"https://railway.com/pricing.md"},{"id":"railway-buckets","provider":"railway","name":"Buckets","category":"object-storage","tagline":"Store uploads and files with no fee to download them.","description":"Buckets are S3-compatible storage, which means they work with tools built for Amazon's S3, a widely used file storage service. If your app lets people upload photos or documents, this is where you keep them. It is a newer product on Railway. The big draw is free egress, meaning no charge when files are downloaded or shown to your users, so it suits apps that serve lots of files. Storage costs $0.015 per GB each month.","useCases":["Store uploaded files cheaply","Serve files without download charges","Connect existing S3-style tools"],"pricing":"$0.015/GB-month, free egress.","status":"ga","url":"https://docs.railway.com/","source":"https://railway.com/pricing.md"},{"id":"railway-cron","provider":"railway","name":"Cron Jobs","category":"workflows","tagline":"Run a task on a schedule, like every night or every hour.","description":"Cron Jobs run one of your Railway services on a schedule you set, using cron, a standard way of writing times like every day at 2am. If you need to send a weekly email digest, clean out old data or refresh a report, this does it without you clicking anything. Pick it over keeping an app running around the clock just to wait for the right time. Billing is by usage.","useCases":["Run a service on a timetable","Run scheduled jobs automatically","Repeat tasks at set times"],"pricing":"Usage-based.","status":"ga","url":"https://docs.railway.com/cron-jobs","source":"https://railway.com/pricing.md"},{"id":"railway-functions","provider":"railway","name":"Railway Functions","category":"serverless","tagline":"Run small pieces of TypeScript code, like handling webhooks.","description":"Railway Functions let you run small snippets of TypeScript, a typed version of JavaScript, right inside your project. If another service needs to notify your app when something happens, like a payment going through, a function can catch that message (called a webhook) without building a full app around it. Pick it for small jobs where setting up a whole service feels like overkill. It is in beta, so details may still change, and it is worth checking current availability before you plan around it. Billing is by usage.","useCases":["Handle incoming webhooks","Run small TypeScript functions","Run short code snippets"],"pricing":"Usage-based.","status":"beta","url":"https://docs.railway.com/functions","source":"https://railway.com/pricing.md"},{"id":"railway-private-networking","provider":"railway","name":"Private Networking and Environments","category":"networking","also":["devtools"],"tagline":"Keep services talking privately, and get a test copy per code change.","description":"Private Networking and Environments gives the parts of your Railway project an encrypted internal network, so they can talk to each other without going over the public internet. It also creates ephemeral (temporary) environments, such as one for each pull request, which is a proposed change to your code. If you want to check a change on a live copy of your app before it ships, this is how. Pick it to keep things like your database away from public access. It is included in Railway's pricing.","useCases":["Connect services over a private network","Spin up preview apps per change","Keep internal traffic encrypted"],"pricing":"Included.","status":"ga","url":"https://docs.railway.com/networking/private-networking","source":"https://railway.com/pricing.md"},{"id":"backblaze-b2","provider":"backblaze","name":"B2 Cloud Storage","category":"object-storage","tagline":"Cheap online storage for files and backups, with a free download allowance.","description":"B2 Cloud Storage is Backblaze's main service for keeping files online, priced at $6.95 per terabyte each month. If your app lets people upload photos or videos, or you need a safe place for nightly backups, you can park them here instead of on your own server. It works with S3-compatible tools, meaning software built for Amazon's popular storage service can usually connect to it. Pick it over the big-name clouds when cost matters, since it is known for low prices and a generous free allowance for downloading your data back out.","useCases":["Store backups at low cost","Archive media files long term","Serve files with free download allowance"],"pricing":"$6.95/TB-month.","status":"ga","url":"https://www.backblaze.com/cloud-storage","source":"https://www.backblaze.com/"},{"id":"backblaze-b2-overdrive","provider":"backblaze","name":"B2 Overdrive","category":"object-storage","tagline":"Fast storage built for very large datasets, such as AI training data.","description":"B2 Overdrive is a version of Backblaze's file storage aimed at moving very large amounts of data quickly. If you are training an AI model and need to feed it huge piles of images or text, or you manage a data lake (one big pool of raw data), this is the kind of job it targets. Choose it over regular B2 Cloud Storage when speed on giant datasets matters more than a simple price list. Pricing is not published, so you will need to contact Backblaze sales.","useCases":["Store large datasets for AI","Feed data to AI training","Manage data lakes at high speed"],"pricing":"Contact sales.","status":"ga","url":"https://www.backblaze.com/cloud-storage/b2-overdrive","source":"https://www.backblaze.com/"},{"id":"backblaze-b2-neo","provider":"backblaze","name":"B2 Neo (B2 for Neoclouds)","category":"object-storage","tagline":"Storage packaged for companies that rent out GPU computing.","description":"B2 Neo, also called B2 for Neoclouds, is Backblaze storage packaged for companies that rent out GPU servers, the chips that run AI work. A neocloud is a newer kind of cloud company focused on that. If you are building such a platform and need somewhere to keep your customers' data, this gives you storage with enterprise features. Most app builders will be better served by standard B2 Cloud Storage. Pricing is handled by contacting Backblaze sales.","useCases":["Store files for GPU cloud platforms","Add enterprise storage to GPU clouds","Offer storage to GPU cloud customers"],"pricing":"Contact sales.","status":"ga","url":"https://www.backblaze.com/cloud-storage","source":"https://www.backblaze.com/"},{"id":"backblaze-computer-backup","provider":"backblaze","name":"Computer Backup","category":"block-storage","also":["end-user"],"tagline":"Unlimited backup for your Mac or PC, from $9 a month.","description":"Computer Backup saves a copy of everything on your laptop or desktop to the cloud, so a lost or broken machine does not mean lost work. If you are a founder with all your files, designs and notes on one laptop, this protects them. It covers Mac and PC with unlimited storage per computer, on both personal and business plans. It is for backing up your own machines, not for storing files your app serves to users; for that, look at B2 Cloud Storage. Plans start at $9 a month per computer.","useCases":["Back up a laptop or desktop","Back up a Mac or PC","Protect work computers with backups"],"pricing":"From $9/month per computer.","status":"ga","url":"https://www.backblaze.com/cloud-backup","source":"https://www.backblaze.com/"},{"id":"backblaze-partner-api","provider":"backblaze","name":"Backblaze Partner API","category":"object-storage","tagline":"Lets partners resell and manage B2 storage and backups from their own code.","description":"Backblaze Partner API is a way for your software to set up and manage B2 storage and backup accounts on behalf of other customers. If you run an agency or software business and want to sell storage or backups under your own service, this is the tool for it. It is meant for resellers, not for people who just need storage for their own app. It is not listed on Backblaze's home page, so confirm with Backblaze that it is available before you build on it. Pricing is by contacting sales.","useCases":["Manage B2 storage through code","Resell storage and backups to customers","Automate partner backup accounts"],"pricing":"Contact sales.","status":"ga","url":"https://www.backblaze.com/cloud-storage","source":"https://www.backblaze.com/"},{"id":"backblaze-universal-data-migration","provider":"backblaze","name":"Universal Data Migration","category":"migration","tagline":"Backblaze moves your data from other clouds into B2 for free.","description":"Universal Data Migration is a free service where Backblaze moves your existing files into B2 for you. If your photos, videos or backups currently sit in Amazon S3 or another cloud and you want out, you do not have to write transfer scripts or work out the move on your own. It is aimed at people leaving S3 for cheaper storage. Pick it when the hassle of moving a lot of data is the thing keeping you where you are.","useCases":["Move data into B2 for free","Switch storage from other clouds","Have Backblaze move your data"],"pricing":"Free service.","status":"ga","url":"https://www.backblaze.com/cloud-storage/solutions/cloud-migration","source":"https://www.backblaze.com/"},{"id":"wasabi-hot-cloud-storage","provider":"wasabi","name":"Wasabi Hot Cloud Storage","category":"object-storage","tagline":"Online file storage with no fees to download your data or make requests.","description":"Wasabi Hot Cloud Storage is a place to keep files online for one flat price, about $6.99 per terabyte each month, with a 1 terabyte minimum. If your app stores user videos or you need a destination for backups, you will not see surprise bills for downloading files or making requests. It works with S3-compatible tools, meaning software built for Amazon's storage service can usually connect to it. Pick it over the big clouds when predictable costs matter more than extras.","useCases":["Use as a backup target","Store media files with flat pricing","Pay no egress or API fees"],"pricing":"About $6.99/TB-month.","status":"ga","url":"https://wasabi.com/hot-cloud-storage","source":"https://www.wasabi.com/"},{"id":"wasabi-cyber-resilient","provider":"wasabi","name":"Cyber-Resilient Storage","category":"object-storage","also":["block-storage"],"tagline":"Lock files so ransomware or mistakes cannot delete or change them.","description":"Cyber-Resilient Storage is a set of protections on Wasabi that make stored files hard to wipe out. Object Lock stops files from being changed or deleted for a set time. Covert Copy keeps a separate protected copy, and Multi-User Authorization requires more than one person to approve risky actions. If ransomware (malware that locks up your data and demands payment) hit your team, you could restore clean copies. It is listed as included with Wasabi storage.","useCases":["Protect files from ransomware","Lock files against changes","Require multiple users to approve changes"],"pricing":"Included.","status":"ga","url":"https://www.wasabi.com/","source":"https://www.wasabi.com/"},{"id":"wasabi-ai-storage","provider":"wasabi","name":"Wasabi AI Storage","category":"object-storage","tagline":"Flat-priced storage for AI training data and model checkpoints.","description":"Wasabi AI Storage is Wasabi's file storage positioned for the huge piles of data that AI work produces, up to petabyte scale (a petabyte is a thousand terabytes). If you are training a model, you need room for training data and for checkpoints, which are saved snapshots of your progress. Pricing is the same flat rate as regular Wasabi storage, so there is no separate AI price to work out. Choose it when you want AI storage costs you can predict.","useCases":["Store training datasets at scale","Keep model checkpoints in storage","Store petabytes of AI data"],"pricing":"Same flat pricing.","status":"ga","url":"https://www.wasabi.com/","source":"https://www.wasabi.com/"},{"id":"wasabi-reserved-capacity","provider":"wasabi","name":"Reserved Capacity Storage","category":"object-storage","tagline":"Pay up front for a set amount of storage at a lower rate.","description":"Reserved Capacity Storage lets you commit to a set amount of Wasabi storage in advance and pay a lower rate in return. If you know your archive of customer videos or old backups will take roughly the same space for a long time, this can cost less than paying as you go. It makes the most sense when your storage needs are steady and predictable. This option is not confirmed on Wasabi's home page, so check the details with Wasabi sales, who handle pricing.","useCases":["Pay upfront for predictable storage","Keep large archives at lower cost","Plan storage needs in advance"],"pricing":"Contact sales.","status":"ga","url":"https://wasabi.com/pricing","source":"https://www.wasabi.com/"},{"id":"wasabi-cloud-nas","provider":"wasabi","name":"Wasabi Cloud NAS","category":"block-storage","tagline":"Shared network drives for your team, stored on Wasabi.","description":"Wasabi Cloud NAS gives you cloud file shares, meaning shared folders your team can open like a drive on their computer, with the files kept in Wasabi storage. NAS stands for network attached storage. If your team passes around big design files or footage and you want one shared drive without managing hardware, this fits. It is for shared team folders, not for files your app serves to users. It is not confirmed on Wasabi's home page and is priced as an add-on, so check before relying on it.","useCases":["Share files across a team","Run shared drives in the cloud","Keep office files on cloud storage"],"pricing":"Add-on pricing.","status":"ga","url":"https://wasabi.com/cloud-nas","source":"https://www.wasabi.com/"},{"id":"wasabi-surveillance-cloud","provider":"wasabi","name":"Wasabi Surveillance Cloud","category":"video","tagline":"Online storage for security camera video.","description":"Wasabi Surveillance Cloud is storage built for video from security cameras. If you run a shop, office or property and need to keep CCTV footage for weeks or months, you can send recordings here rather than filling up local drives. It builds on Wasabi's cloud storage, so it suits keeping a lot of footage for a long time. It is not aimed at app builders, so skip it unless cameras are involved. It is not confirmed on Wasabi's home page, and pricing is through sales.","useCases":["Keep security camera footage on file","Retain CCTV recordings for review","Store video from surveillance systems"],"pricing":"Contact sales.","status":"ga","url":"https://wasabi.com/surveillance","source":"https://www.wasabi.com/"},{"id":"supabase-database","provider":"supabase","name":"Supabase Database","category":"sql","tagline":"A full Postgres database for your app, with ready-made APIs.","description":"Supabase Database gives every project its own Postgres, a popular and well-trusted database, with REST and GraphQL APIs (ready-made ways for your app to read and save data). If you are building a SaaS product that keeps customers, orders and settings, this is where they live. It includes more than 40 extensions (add-ons that teach the database new tricks), read replicas (extra copies for faster reading) and branching. Pick it when you want standard Postgres you can take elsewhere later. There is a free tier, and Pro starts at $25 per month.","useCases":["Power the data behind an app","Store records for a SaaS product","Read and write data over web APIs"],"pricing":"Free tier; Pro from $25/month.","status":"ga","url":"https://supabase.com/database","source":"https://supabase.com/features"},{"id":"supabase-auth","provider":"supabase","name":"Supabase Auth","category":"identity","tagline":"Let people sign up and log in with email, phone or social accounts.","description":"Supabase Auth handles sign up and login for your app, so you do not have to build it yourself. If you are making a members-only product, people can log in with email, phone, magic links (a one-click link sent to their inbox) or more than 20 social providers. It also supports SSO (single sign-on, where staff use their company account). Because it ties into your Supabase database, you can use RLS, rules that decide which user sees which rows. The first 50,000 monthly active users are free, then pricing follows usage.","useCases":["Let users sign up and log in","Log users in with email or phone","Offer single sign-on for companies"],"pricing":"50k MAU free; then usage.","status":"ga","url":"https://supabase.com/auth","source":"https://supabase.com/features"},{"id":"supabase-storage","provider":"supabase","name":"Supabase Storage","category":"object-storage","tagline":"Store user uploads like photos and files, with built-in image resizing.","description":"Supabase Storage keeps the files your users upload, such as photos, documents and videos. If your app lets people upload a profile picture, it can store that file, serve it quickly through a CDN (a network that delivers files from a location near each visitor) and transform the image, for example by resizing it. It is S3-compatible, meaning tools built for Amazon's file storage can talk to it. It offers three bucket types (a bucket is a folder-like container): files, analytics and vector. You get 1 GB free, then pay by usage.","useCases":["Store files users upload","Resize images as they load","Keep files delivered quickly worldwide"],"pricing":"1 GB free; usage-based.","status":"ga","url":"https://supabase.com/storage","source":"https://supabase.com/features"},{"id":"supabase-edge-functions","provider":"supabase","name":"Edge Functions","category":"serverless","also":["edge-compute"],"tagline":"Run small pieces of backend code close to your users.","description":"Edge Functions let you run small bits of server code without managing a server. You write them in TypeScript (JavaScript with extra safety checks) on Deno, a modern runtime, and they run at the edge, meaning in data centers near your users. If you need to receive a webhook (a message another service sends when something happens, like a payment) or call an AI service from your app, this is a good fit. It supports Node compatibility and regional invocation, so you can choose where a function runs. Included allowances come with your plan.","useCases":["Run webhook handlers near users","Call AI models from server code","Run backend code close to users"],"pricing":"Included allowances.","status":"ga","url":"https://supabase.com/edge-functions","source":"https://supabase.com/features"},{"id":"supabase-realtime","provider":"supabase","name":"Supabase Realtime","category":"realtime-db","also":["realtime"],"tagline":"Show live updates in your app, like chat messages and changing numbers.","description":"Supabase Realtime pushes changes to your users' screens the moment they happen, with no refreshing. If you are building a chat feature or a live dashboard, it can broadcast messages to everyone in a room, show who is online (Presence), and send out database changes as they happen (Postgres Changes). It does this over WebSockets, a kind of always-open connection between your app and the server. It is built on Elixir, a language known for handling many connections at once. It is included with your plan, within quotas.","useCases":["Build live chat between users","Update dashboards as data changes","Show who is online now"],"pricing":"Included with quotas.","status":"ga","url":"https://supabase.com/realtime","source":"https://supabase.com/features"},{"id":"supabase-vector","provider":"supabase","name":"Supabase Vector","category":"vector","tagline":"Store AI embeddings next to your data, inside Postgres.","description":"Supabase Vector uses pgvector, an add-on that lets Postgres store embeddings, which are lists of numbers that capture the meaning of text or images. If you are building a chatbot that answers questions from your own documents (often called RAG) or a recommendations feature, you can search by meaning instead of exact words. Because the embeddings sit with the rest of your data, you do not need a second database to look after. Pick it when your data is already in Supabase. It is part of the database, so there is no separate price.","useCases":["Search content by meaning","Power AI retrieval with your data","Suggest related products to shoppers"],"pricing":"Part of database.","status":"ga","url":"https://supabase.com/modules/vector","source":"https://supabase.com/features"},{"id":"supabase-vector-buckets","provider":"supabase","name":"Vector Buckets","category":"vector","tagline":"Early access storage for very large sets of AI embeddings.","description":"Vector Buckets store embeddings, lists of numbers that capture meaning, in file-style storage backed by S3 (Amazon's file storage), and let you search for similar items. If you have a large collection of embeddings from your documents or products, this gives you a place to keep them outside your main database. It is worth a look when your embeddings outgrow what you want to keep alongside your regular app data. It is in Public Alpha, meaning it is early and may change, and it uses alpha pricing.","useCases":["Store large sets of embeddings","Run similarity search at scale","Keep AI search data in storage"],"pricing":"Alpha pricing.","status":"preview","url":"https://supabase.com/features","source":"https://supabase.com/features"},{"id":"supabase-analytics-buckets","provider":"supabase","name":"Analytics Buckets","category":"warehouse","tagline":"Early access storage for analyzing your app's data in bulk.","description":"Analytics Buckets keep your data in Iceberg tables, an open format for large piles of data you want to analyze, inside Supabase Storage. If you want to run reports on the data your app has collected in Postgres, this gives you a place for a copy that analysis tools can read. It suits analysis work, rather than the day to day reading and saving your app does. It is in Public Alpha, meaning it is early and may change, and it uses alpha pricing.","useCases":["Store tables for analytics work","Analyze data kept in Postgres","Run reports on large datasets"],"pricing":"Alpha pricing.","status":"preview","url":"https://supabase.com/features","source":"https://supabase.com/features"},{"id":"supabase-queues","provider":"supabase","name":"Supabase Queues","category":"queues","tagline":"A to-do list for background jobs, kept safely in Postgres.","description":"Supabase Queues is a waiting line for tasks your app does in the background, built on pgmq, a Postgres add-on. If someone signs up and you want to send a welcome email without making them wait, you drop the task in the queue and a worker picks it up later. The queue is durable, meaning tasks are saved and not lost if something restarts. You do not need a separate queue service. It is in Public Alpha, so expect changes, and it is included in your plan.","useCases":["Run background jobs in order","Queue tasks like emails for later","Hold work until it is processed"],"pricing":"Included.","status":"preview","url":"https://supabase.com/modules/queues","source":"https://supabase.com/features"},{"id":"supabase-cron","provider":"supabase","name":"Supabase Cron","category":"workflows","tagline":"Run tasks on a schedule, like nightly cleanups.","description":"Supabase Cron runs jobs on a timetable, like an alarm clock for your database. If you want to delete old sign-in records every night, you set the schedule once and it runs on its own. It wraps pg_cron, a well-known Postgres add-on for scheduling, so the work happens right where your data lives. Pick it when the job is about your database and you do not want another tool to manage. It is in beta, meaning it is still being polished, and it is included.","useCases":["Run cleanup jobs on a schedule","Refresh data every night automatically","Schedule database tasks on a timer"],"pricing":"Included.","status":"beta","url":"https://supabase.com/modules/cron","source":"https://supabase.com/features"},{"id":"supabase-branching","provider":"supabase","name":"Database Branching","category":"devtools","tagline":"Test database changes safely in a separate copy for each pull request.","description":"Database Branching gives you an isolated Postgres for each pull request, which is a proposed code change waiting for review. If you are changing your app's tables, you can try the change on a preview branch, along with its migrations (saved step by step database changes), without touching real customer data. Teammates can check the preview before anything goes live. Pick it when more than one person is changing the app and you want mistakes to happen in a copy. Pricing is usage-based.","useCases":["Preview each pull request safely","Test database changes before launch","Give each change its own database"],"pricing":"Usage-based.","status":"ga","url":"https://supabase.com/docs/guides/deployment/branching","source":"https://supabase.com/features"},{"id":"supabase-pipelines","provider":"supabase","name":"Supabase Pipelines","category":"data-pipelines","tagline":"Copy your Postgres data into BigQuery automatically, in early access.","description":"Supabase Pipelines keeps a copy of your Postgres data flowing into BigQuery, Google's tool for analyzing large amounts of data. It uses CDC (change data capture), which means it sends only what changed instead of copying everything again. If your team wants charts and reports on product data, this is a managed way to feed them. It has been in public alpha since July 21, 2026, on paid plans.","useCases":["Sync Postgres data to BigQuery","Keep analytics data current","Stream database changes to analytics"],"pricing":"Alpha.","status":"preview","url":"https://supabase.com/changelog","source":"https://supabase.com/features"},{"id":"supabase-vault","provider":"supabase","name":"Supabase Vault","category":"secrets","tagline":"Keep API keys and other secrets stored safely inside your database.","description":"Supabase Vault is a spot inside your Supabase database (Postgres, a popular database) where you can keep secrets, like the passwords and API keys your app uses to talk to other services. If your app connects to a payment or email service, you can store that key here instead of pasting it into your code. It keeps the secret next to your data, so there is one less tool to set up. It is listed as a public alpha, which means it is early and still changing, so expect rough edges.","useCases":["Store API keys securely","Keep secrets out of code","Save private credentials in the database"],"pricing":"Included.","status":"preview","url":"https://supabase.com/features","source":"https://supabase.com/features"},{"id":"supabase-read-replicas","provider":"supabase","name":"Read Replicas","category":"sql","tagline":"Extra copies of your database that handle reading, so your app stays fast.","description":"Read Replicas are extra copies of your Supabase database that only answer questions, such as loading a page of data, while the main database handles changes. If lots of people browsing your app slow things down for everyone, you can send that browsing traffic to a replica. You can also place copies closer to faraway users so pages load faster for them. Pick this over simply buying a bigger database when the slowdown comes from many people reading at once. You pay for the server power each replica uses.","useCases":["Spread heavy read traffic out","Serve readers closer to users","Speed up apps with many readers"],"pricing":"Per replica compute.","status":"ga","url":"https://supabase.com/docs/guides/platform/read-replicas","source":"https://supabase.com/features"},{"id":"supabase-logs-analytics","provider":"supabase","name":"Logs and Observability","category":"observability","tagline":"See what your app and database are doing, so you can find what broke.","description":"Logs and Observability gives you a place to search the running record of what your Supabase project is doing, plus reports that summarize it. If a signup fails or a page loads slowly, you can look through the logs to see what happened and when. It is built into Supabase, so you do not need to connect a separate monitoring tool just to start debugging. If you run Supabase on your own servers, it is available as an opt-in since June 2026. It is included in the service.","useCases":["Find errors in app logs","Review reports on app activity","Debug problems faster"],"pricing":"Included.","status":"ga","url":"https://supabase.com/docs/guides/telemetry/logs","source":"https://supabase.com/features"},{"id":"neon-postgres","provider":"neon","name":"Neon Postgres (Lakebase Postgres)","category":"sql","also":["nosql"],"tagline":"A database that grows with traffic and switches off when nobody is using it.","description":"Neon Postgres is a hosted Postgres, a popular database, that grows and shrinks on its own as your app gets busier or quieter. If your app saves users, orders or posts, use it as your main database, or make a separate copy for each change you want to test. It can drop to zero when nobody is using it, which suits side projects. Neon is now part of Databricks and calls its database Lakebase Postgres. There is a free plan, and Launch is pay as you go from $0.106 per CU-hour (one unit of computing power for an hour).","useCases":["Power app databases that scale","Create a database for each change","Pause the database when unused"],"pricing":"Free plan; Launch usage-based.","status":"ga","url":"https://neon.com/","source":"https://neon.com/docs"},{"id":"neon-branching","provider":"neon","name":"Database Branching","category":"devtools","tagline":"Make instant copies of your database to test changes safely.","description":"Database Branching lets you make a copy of your whole Neon database in moments, without waiting for all the data to be duplicated. If you want to try a risky change, like reshaping how your data is stored, you can try it on a branch and leave your real data alone. It also works well for giving every preview version of your app its own database. Branches only store what you change (called copy-on-write), which is why they are quick to create. Launch includes up to 10 branches per project, and extra ones cost $1.50 per branch-month.","useCases":["Make instant copies for testing","Spin up a database per preview","Test changes on copies of data"],"pricing":"$1.50/extra branch-month.","status":"ga","url":"https://neon.com/docs","source":"https://neon.com/docs"},{"id":"neon-auth","provider":"neon","name":"Managed Better Auth","category":"identity","also":["sql"],"tagline":"Ready-made sign-up and login for your app, run for you.","description":"Managed Better Auth gives your app sign-up and login without you building them yourself. Better Auth is a popular login toolkit, and here it is run for you as a managed service. If your app needs people to create accounts and come back later, this covers it, and it comes from the same provider as your database. It was previously called Neon Auth. It is included, with 60k MAU (people who use your app in a month) free and up to 1M MAU on paid plans.","useCases":["Let users sign up and log in","Add user accounts to a product","Manage logins without building auth"],"pricing":"Included.","status":"ga","url":"https://neon.com/docs","source":"https://neon.com/docs"},{"id":"neon-data-api","provider":"neon","name":"Neon Data API","category":"sql","tagline":"Let your website talk straight to your database over the web.","description":"Neon Data API lets your app read and write your Postgres database (a popular database) through ordinary web requests, in the style of PostgREST, a tool that turns a database into a web interface. If you are building a simple front end, like a to-do app, it can ask for data directly without you writing a separate backend first. It works with RLS (row-level security), which are rules in the database that decide who can see which rows. Pick it when you want to move fast and keep access rules next to your data. It is included.","useCases":["Let frontend code query data directly","Read and write data over HTTP","Skip building a custom API layer"],"pricing":"Included.","status":"ga","url":"https://neon.com/docs","source":"https://neon.com/docs"},{"id":"neon-functions","provider":"neon","name":"Neon Functions","category":"serverless","tagline":"Run your backend code right next to your database, with no time limit.","description":"Neon Functions runs small pieces of your backend code, the behind-the-scenes logic of your app, close to your database. If you need to work out an order total or process data whenever something happens in your app, you can put that code here. Because the code sits near your data and has no timeouts, longer jobs are less likely to get cut off. Pricing has two parts: $0.60 per million invocations (times your code runs), plus charges for capacity-hours, meaning how much computing power you use over time.","useCases":["Run backend logic close to data","Run server code without timeouts","Handle requests next to the database"],"pricing":"$0.60/M invocations plus capacity.","status":"ga","url":"https://neon.com/docs","source":"https://neon.com/docs"},{"id":"neon-object-storage","provider":"neon","name":"Neon Object Storage","category":"object-storage","tagline":"Store uploads and images in a place that branches along with your project.","description":"Neon Object Storage is a place to keep files, like photos, documents and other uploads, right alongside your Neon project. If your app lets people upload profile pictures, this is where those files can live. It is S3-compatible, meaning it speaks the same language as Amazon's popular file storage, so tools built for that can connect to it. What sets it apart is that your files branch together with your project, so test copies get their own set. The first 5 GB are free, then it is $0.023 per GB-month on Launch.","useCases":["Store user uploaded files","Keep files alongside database branches","Store documents and images for apps"],"pricing":"5 GB free.","status":"ga","url":"https://neon.com/docs","source":"https://neon.com/docs"},{"id":"neon-ai-gateway","provider":"neon","name":"Neon AI Gateway","category":"ai-gateway","tagline":"One connection to many AI models, built into your project.","description":"Neon AI Gateway gives your project a single way to use many AI models, both top commercial ones and open ones, powered by Databricks. If you are adding a chat box or text summaries to your app, you connect to one place instead of signing up with each AI company. It speaks the OpenAI format and the Gemini format, which are what many AI tools expect, so tools built for them can connect. You pay with prepaid credits that you buy ahead of time.","useCases":["Call many AI models through one API","Add chat and text features to apps","Use OpenAI-style code with other models"],"pricing":"Prepaid credits.","status":"ga","url":"https://neon.com/docs","source":"https://neon.com/docs"},{"id":"neon-realtime","provider":"neon","name":"Neon Realtime","category":"realtime-db","also":["realtime"],"tagline":"Show live changes in your app the moment your data changes.","description":"Neon Realtime sends updates from your Postgres database (a popular database) to your app as they happen, so screens refresh without anyone reloading the page. If you are building a chat, a live dashboard or a shared to-do list, this keeps everyone looking at the latest information. Pick it when you want live updates straight from the data you already store, instead of adding a separate live-update service. It is in beta, which means it is still being tested, and it is free during the beta.","useCases":["Push live updates to the interface","Show database changes as they happen","Keep screens in sync with data"],"pricing":"Free during beta.","status":"beta","url":"https://neon.com/pricing","source":"https://neon.com/docs"},{"id":"neon-instant-restore","provider":"neon","name":"Instant Restore and Snapshots","category":"block-storage","tagline":"Rewind your database to an earlier moment if something goes wrong.","description":"Instant Restore and Snapshots lets you roll your Neon database back to an earlier point in time, save snapshots (saved copies of your data), and set up scheduled backups. If a bad update wipes out customer records, you can go back to just before it happened. It covers you without needing to remember to make manual backups. The Scale plan keeps up to 30 days of history. Storing that history costs $0.20 per GB-month.","useCases":["Undo mistakes with point-in-time restore","Keep snapshots of your data","Run scheduled backups automatically"],"pricing":"$0.20/GB-month history.","status":"ga","url":"https://neon.com/pricing","source":"https://neon.com/docs"},{"id":"neon-agent-plan","provider":"neon","name":"Agent Plan","category":"ai-platform","tagline":"Neon pricing for apps that create thousands of databases for AI agents.","description":"Agent Plan is a Neon pricing plan for products that create databases automatically, sometimes thousands of them, for AI agents. It gives you unlimited projects (each project is its own database setup) at the rates of Neon's Launch plan, plus free-tier credits. If you run an app builder where every new user or agent gets a fresh database, the number of databases you create is not capped. Pick it over a standard plan when your growth comes from database count rather than database size. Pricing starts at $0.106 per CU-hour, a unit of database computing time.","useCases":["Create databases for many AI agents","Give each app builder its own database","Provision thousands of databases at scale"],"pricing":"From $0.106/CU-hour.","status":"ga","url":"https://neon.com/programs/agents","source":"https://neon.com/docs"},{"id":"planetscale-postgres","provider":"planetscale","name":"PlanetScale Postgres","category":"sql","tagline":"Postgres, the popular database, managed for you from $5 a month.","description":"PlanetScale Postgres is Postgres, a popular database, run for you so you don't have to look after servers yourself. If you're building a SaaS product that needs to store customer accounts and their data, this is a place to keep it. You can start with a single node (one server) from $5 a month, then move to a high availability setup from $15 a month. That setup has a primary copy backed by two replicas, which are live spare copies that can step in if something fails. It has been generally available since 2025, and you can choose EBS or Metal storage.","useCases":["Run a managed Postgres database","Host databases for a SaaS product","Start on a single small server"],"pricing":"Single node from $5; HA from $15/month.","status":"ga","url":"https://planetscale.com/postgres","source":"https://planetscale.com/pricing"},{"id":"planetscale-vitess","provider":"planetscale","name":"PlanetScale Vitess (MySQL)","category":"sql","tagline":"MySQL that can grow across many servers, from $39 a month.","description":"PlanetScale Vitess runs MySQL, a widely used database, on Vitess, a system that can split one large database across many servers. That splitting is called sharding. If your MySQL database is getting too big or too busy for a single machine, this is built for that moment. It also supports non-blocking schema changes, which means you can change how your tables are laid out without freezing your app. Choose it over PlanetScale Postgres when your app already speaks MySQL or you expect very large scale. High availability plans start at $39 a month.","useCases":["Run large MySQL databases","Split data across many servers","Change table layouts without stopping traffic"],"pricing":"HA from $39/month.","status":"ga","url":"https://planetscale.com/mysql","source":"https://planetscale.com/pricing"},{"id":"planetscale-metal","provider":"planetscale","name":"PlanetScale Metal","category":"sql","tagline":"Databases on very fast local drives, for apps with heavy traffic.","description":"PlanetScale Metal runs your database on NVMe, a very fast type of storage attached directly to the server. It works for both Postgres and Vitess (MySQL) clusters. Because the drives are local, you get unlimited IOPS, which means no cap on how many reads and writes per second the storage handles. If your app handles a constant stream of small actions, like orders, messages or clicks, that speed helps. Pick it over standard storage when your database is the busiest part of your product. High availability setups start at $50 a month.","useCases":["Handle heavy read and write traffic","Speed up busy databases with fast disks","Run high-throughput Postgres or MySQL"],"pricing":"HA from $50/month.","status":"ga","url":"https://planetscale.com/metal","source":"https://planetscale.com/pricing"},{"id":"planetscale-neki","provider":"planetscale","name":"Neki","category":"sql","tagline":"Postgres that spreads across many servers, from the Vitess team.","description":"Neki is sharded Postgres, which means Postgres, a popular database, split across many servers so it can keep growing. It comes from the team behind Vitess, the technology PlanetScale uses to scale MySQL. If you think your Postgres database could one day outgrow a single machine, Neki is aimed at that problem. Until then, regular PlanetScale Postgres is the simpler choice. Neki is in preview, it appears in PlanetScale's plan docs, public availability is unconfirmed, and pricing has not been published.","useCases":["Split Postgres data across servers","Grow Postgres beyond one server","Try an early preview of sharded Postgres"],"pricing":"Not published.","status":"preview","url":"https://planetscale.com/docs/planetscale-plans","source":"https://planetscale.com/pricing"},{"id":"planetscale-insights","provider":"planetscale","name":"Query Insights","category":"observability","tagline":"See which database requests are slow, built into every database.","description":"Query Insights shows you how each request your app sends to the database is performing. If your app suddenly feels slow, you can use it to spot which query, meaning a single request for data, is taking the longest. It is built into every PlanetScale database, so there is nothing extra to install or connect. Pick it when you want a quick answer to why things are slow without adding a separate monitoring tool. It is included at no extra cost.","useCases":["Find slow database queries","See performance stats for each query","Spot what slows the app down"],"pricing":"Included.","status":"ga","url":"https://planetscale.com/features/insights","source":"https://planetscale.com/pricing"},{"id":"planetscale-branching","provider":"planetscale","name":"Database Branching and Deploy Requests","category":"devtools","tagline":"Try database changes on a copy, then push them live safely.","description":"Database Branching lets you make a copy of your database's structure, called a branch, and try changes there before touching the real thing. When you're happy, you send a deploy request, which is a proposal to apply those changes to your live database safely. If you need to add a new column to a table while customers are using your app, this lets you do it without crossing your fingers. It comes from Vitess, the technology behind PlanetScale's MySQL. Your plan includes 1440 branch hours, which count how long your branches stay running.","useCases":["Test schema changes in a copy","Ship database changes safely","Create test copies of the database"],"pricing":"1440 branch hours included.","status":"ga","url":"https://planetscale.com/docs/vitess/schema-changes/branching","source":"https://planetscale.com/pricing"},{"id":"planetscale-traffic-control","provider":"planetscale","name":"Traffic Control","category":"management","tagline":"Slow down greedy database requests before they overload your app.","description":"Traffic Control lets you slow down database queries, meaning requests for data, that are using more than their share. Imagine a bug or a heavy report suddenly floods your database and every customer feels it. With Traffic Control you can throttle the noisy requests so the rest of your app keeps working. It protects your cluster, which is your set of database servers, from runaway workloads. It is included on supported plans, so check that yours is one of them.","useCases":["Throttle runaway database queries","Protect the database from overload","Keep heavy jobs from crashing the app"],"pricing":"Included on supported plans.","status":"ga","url":"https://planetscale.com/docs","source":"https://planetscale.com/pricing"},{"id":"turso-cloud","provider":"turso","name":"Turso Cloud","category":"sql","tagline":"Hosted SQLite databases, with up to 100 on the free plan.","description":"Turso Cloud hosts SQLite, a lightweight and very popular database, so you don't have to run it yourself. It is built to hold huge numbers of separate databases on AWS. If you want one database per customer, or a small database to hold each AI agent's memory and state, this fits well. Pick it over a traditional single big database when your product naturally splits into many small ones. The free plan allows 100 databases. Paid plans are Developer at $4.99, Scaler at $24.92 and Pro at $416.58.","useCases":["Give each customer their own database","Store state for AI agents","Host many small SQLite databases"],"pricing":"Free; Developer $4.99; Scaler $24.92; Pro $416.58.","status":"ga","url":"https://turso.tech/","source":"https://turso.tech/pricing"},{"id":"turso-database","provider":"turso","name":"Turso Database","category":"sql","tagline":"A rewrite of SQLite that runs inside your app, now in beta.","description":"Turso Database is a fresh rewrite of SQLite, a lightweight and popular database, written in Rust, a modern programming language. It was formerly called Limbo. It runs in-process, meaning inside your app itself rather than as a separate server. It stays compatible with SQLite while adding async I/O, which lets it do several things at once, and vector search, which finds items by meaning rather than exact words. If you're building an embedded or desktop-style app, it's worth a look. It is free and open source, and it is currently in beta.","useCases":["Build apps with an embedded database","Use a SQLite-compatible database in the app","Try async and vector features in beta"],"pricing":"Free, open source.","status":"beta","url":"https://github.com/tursodatabase/turso","source":"https://turso.tech/pricing"},{"id":"turso-libsql","provider":"turso","name":"libSQL","category":"sql","tagline":"A free, open source spin-off of SQLite you can run yourself.","description":"libSQL is a fork of SQLite, meaning a separate copy of the code that is developed on its own. SQLite is a lightweight, popular database. If you want to run SQLite on your own servers instead of paying for a hosted service, libSQL is free and open source, so you can do that. It has been proven in production. Choose it when you want to self-host and stay in control. Note that Turso now focuses its energy on Turso Database, its newer rewrite of SQLite, so that is where most new work is going.","useCases":["Self-host a SQLite database","Run SQLite on your own servers","Use an open source SQLite fork"],"pricing":"Free, open source.","status":"ga","url":"https://github.com/tursodatabase/libsql","source":"https://turso.tech/pricing"},{"id":"turso-sync","provider":"turso","name":"Turso Sync and Embedded Replicas","category":"realtime-db","tagline":"Keep a database copy in your app and sync it with the cloud.","description":"Turso Sync and Embedded Replicas let you keep a copy of your database right inside your app and sync it with the cloud. An embedded replica is simply that local copy. If you're building an offline-first app, like a notes or field-work tool that must keep working without internet, your users can keep reading and writing, and their changes sync when they reconnect. Choose it over a cloud-only database when speed and offline use matter. Syncs are counted each month on Turso's pricing page, and your plan sets the allowance.","useCases":["Make offline-first apps work without signal","Sync a local database with the cloud","Keep devices' data updated automatically"],"pricing":"Plan allowances.","status":"ga","url":"https://turso.tech/pricing","source":"https://turso.tech/pricing"},{"id":"turso-vector","provider":"turso","name":"Native Vector Search","category":"vector","tagline":"Search by meaning inside your database, with no add-on needed.","description":"Native Vector Search lets your database find things that are similar in meaning, not just ones that match exact words. It is built into Turso, so you don't need to install an extension (an add-on) to get it. If your app turns text or images into embeddings (lists of numbers that capture meaning), you can store them in a column and search them right beside the rest of your data. Pick it over a separate search service when you'd rather keep everything in one place with fewer pieces to manage. It's included at no extra charge.","useCases":["Store embeddings in the main database","Search by meaning using vectors","Add similarity search without extra tools"],"pricing":"Included.","status":"ga","url":"https://turso.tech/vector","source":"https://turso.tech/pricing"},{"id":"turso-branching-pitr","provider":"turso","name":"Branching and Point-in-Time Restore","category":"devtools","tagline":"Make test copies of your database and rewind to an earlier moment.","description":"Branching gives you a separate copy of your database to experiment on, and point-in-time restore lets you rewind to an earlier moment if something goes wrong. If you're about to try a risky change, like reshaping how your data is stored, you can make a branch, test it safely, and throw it away. Both come with no extra usage cost. How far back you can rewind depends on your plan, from 1 day on the free plan to 30 days on enterprise.","useCases":["Restore a database to an earlier time","Spin up test copies of data","Recover from accidental deletes"],"pricing":"Included.","status":"ga","url":"https://turso.tech/pricing","source":"https://turso.tech/pricing"},{"id":"upstash-redis","provider":"upstash","name":"Upstash Redis","category":"kv-cache","tagline":"A very fast data store for temporary data, billed per request.","description":"Upstash Redis is a hosted version of Redis, a popular data store known for speed, and you don't manage any servers. If your app shows the same results to many visitors, you can keep them here (caching) so pages load faster. It's also handy for rate limiting, which means stopping one visitor from hammering your app with too many requests. You can reach it with plain web requests or the standard connection method. There's a free tier, and after that you pay per request, so a quiet app costs very little.","useCases":["Cache slow responses for speed","Limit requests from each user","Store quick key-value data"],"pricing":"Free tier; pay per request.","status":"ga","url":"https://upstash.com/pricing/redis","source":"https://upstash.com/"},{"id":"upstash-vector","provider":"upstash","name":"Upstash Vector","category":"vector","tagline":"Search by meaning for your AI app, with embedding models built in.","description":"Upstash Vector is a database that finds content by meaning rather than exact words, with no servers for you to run. If you're building a chatbot that answers questions from your own documents (often called RAG), you store your text here and look up the most relevant pieces for each question. Its built-in embedding models turn your plain text into searchable numbers for you, so you don't need a separate service for that step. There's a free tier, and after that you pay per request.","useCases":["Power retrieval for AI answers","Store embeddings with built-in models","Search documents by meaning"],"pricing":"Free tier; pay per request.","status":"ga","url":"https://upstash.com/pricing/vector","source":"https://upstash.com/"},{"id":"upstash-search","provider":"upstash","name":"Upstash Search","category":"ai-search","tagline":"Search your site's content by keywords and meaning combined.","description":"Upstash Search is a ready-made search service for the content in your product. If you want a search box on your site or app where people find articles, products or help pages, you can send your content here and query it. It supports full-text search (matching the words people type) and hybrid search, which blends that with matching by meaning, so results can still make sense when someone words things differently. It's a newer addition to Upstash, with a free tier and usage-based pricing after that.","useCases":["Add search to a website","Run full-text and hybrid search","Find content by keywords"],"pricing":"Free tier; usage.","status":"ga","url":"https://upstash.com/","source":"https://upstash.com/"},{"id":"upstash-qstash","provider":"upstash","name":"QStash","category":"queues","tagline":"Send web requests later, on a schedule, or with automatic retries.","description":"QStash is a message queue that works over plain web requests. You hand it a message, and it delivers that message to your app's web address when you say so. If you send webhooks (automatic notifications) to your customers, it can retry when delivery fails, so a short outage doesn't lose anything. It also supports delays, cron (repeating schedules like every morning at 8), and fan-out, which sends one message to many destinations. There's a free tier, and after that you pay per message.","useCases":["Deliver webhooks reliably with retries","Schedule jobs to run on a timer","Send one event to many listeners"],"pricing":"Free tier; pay per message.","status":"ga","url":"https://upstash.com/pricing/qstash","source":"https://upstash.com/"},{"id":"upstash-workflow","provider":"upstash","name":"Upstash Workflow","category":"workflows","tagline":"Run long, multi-step jobs that can pick up where they left off.","description":"Upstash Workflow lets you write a job as a series of steps in a serverless function (code that runs without you managing a server). It's durable, meaning that if something fails partway through, the job can continue instead of starting over. If your app needs to do something long, like generating a report, then emailing it, then updating a record, you can break it into steps here. It's built on QStash, Upstash's message queue, and pricing is based on usage.","useCases":["Run long multi-step background jobs","Keep multi-step processes going after failures","Chain serverless steps reliably"],"pricing":"Usage-based.","status":"ga","url":"https://upstash.com/pricing/workflow","source":"https://upstash.com/"},{"id":"upstash-box","provider":"upstash","name":"Upstash Box","category":"agents","also":["vms"],"tagline":"Cloud computers that stay on, built for your AI agents to work in.","description":"Upstash Box gives an AI agent its own persistent cloud computer, meaning one that keeps its files and state between sessions. If you're building a product where an agent writes code, runs commands or edits files, you can give it a Box as a sandbox, a safe separate space away from your main app. It comes with Claude Code, Codex and OpenCode (AI coding assistants) built in, so you don't have to set them up yourself. Pricing is based on usage.","useCases":["Give AI agents a persistent computer","Run coding agents like Claude Code","Keep agent work sessions going"],"pricing":"Usage-based.","status":"ga","url":"https://upstash.com/box","source":"https://upstash.com/"},{"id":"upstash-blob","provider":"upstash","name":"Upstash Blob","category":"object-storage","tagline":"Store your app's files and images, with a CDN and direct uploads.","description":"Upstash Blob is a place to keep files like photos, videos and documents. If your app lets people upload profile pictures, you can store them here and show them to others quickly. It's S3-compatible, so it works with tools built for Amazon S3, a widely used file storage service. It includes a CDN (a network that serves files from a location near each visitor) and direct uploads, where a visitor's browser sends the file straight to storage. Buckets, the containers that hold your files, can be public or private. Pricing is usage-based.","useCases":["Store user file uploads","Accept direct uploads from browsers","Serve files through a CDN"],"pricing":"Usage-based.","status":"ga","url":"https://upstash.com/blob","source":"https://upstash.com/"},{"id":"upstash-kafka","provider":"upstash","name":"Upstash Kafka","category":"queues","tagline":"Serverless Kafka for streaming data, now retired.","description":"Upstash Kafka was a hosted version of Kafka, a tool for moving large streams of events between different parts of a system. It was used for things like legacy streaming setups, where many systems need to read the same flow of data. If you only need to deliver messages or run scheduled jobs, QStash from the same provider is on the current list. Upstash announced the discontinuation of Kafka in September 2024 and later described it as retired, and it is no longer on their current product list.","useCases":["Legacy message streaming setups","Serverless message streams, now retired","Not offered; product has been retired"],"pricing":"Retired.","status":"closed","url":"https://upstash.com/blog/workflow-kafka","source":"https://upstash.com/"},{"id":"convex-database","provider":"convex","name":"Convex Database","category":"realtime-db","also":["nosql"],"tagline":"A database that updates your app's screens live as data changes.","description":"Convex Database stores your app's data and keeps your screens in sync with it. It is reactive, so when data changes, every screen showing it updates on its own, which Convex does through live queries. If you're building something collaborative, like a shared document or a team task board, you don't have to write code to refresh what people see. You define your data shape in TypeScript (a typed version of JavaScript), and the sync to your app happens automatically. It's free to start, and Pro costs $25 per developer per month.","useCases":["Build collaborative apps with live data","Write data schemas in TypeScript","Sync data to users automatically"],"pricing":"Free; Pro $25/developer/month.","status":"ga","url":"https://www.convex.dev/","source":"https://www.convex.dev/pricing"},{"id":"convex-functions","provider":"convex","name":"Convex Functions","category":"serverless","tagline":"Write your app's backend logic in TypeScript and run it in the cloud.","description":"Convex Functions is where your app's behind-the-scenes logic lives, written in TypeScript (a typed version of JavaScript). Queries read data, mutations change it, and actions can do things like call outside services using Node. If you need rules like \"only let someone join a team if there's a free seat,\" you'd put that here. Queries and mutations are transactional (they fully succeed or don't happen at all) and deterministic (the same input gives the same result), which makes them predictable. The first 1M calls are free, then it's $2.20 per extra million.","useCases":["Run server logic in TypeScript","Update data safely in transactions","Run Node actions that do extra work"],"pricing":"1M calls free; $2.20 per extra million.","status":"ga","url":"https://docs.convex.dev/functions","source":"https://www.convex.dev/pricing"},{"id":"convex-file-storage","provider":"convex","name":"Convex File Storage","category":"object-storage","tagline":"Store files and images right next to your app's data.","description":"Convex File Storage is a built-in place to keep the files your app collects, like images and documents. If your app lets people upload profile photos or attach PDFs, you can save them here without setting up a separate storage account. It sits inside the same backend as the rest of your Convex data, so there is one less service to connect. The first 1 GB is free, and after that it costs $0.033 per GB.","useCases":["Store files people upload","Keep 1 GB free to start","Store uploads inside the same backend"],"pricing":"Usage-based.","status":"ga","url":"https://docs.convex.dev/file-storage","source":"https://www.convex.dev/pricing"},{"id":"convex-auth","provider":"convex","name":"Convex Auth","category":"identity","tagline":"Add sign-in to your app with Clerk, Auth0, or a built-in option.","description":"Convex Auth handles logging people in to your app. If you are building something where users need accounts, you can connect Clerk or Auth0, which are popular sign-in services, or use the library that comes built in. That lets you pick a ready-made service you may already know, or keep everything inside Convex. Authentication is included on all plans, so you do not pay extra to turn it on.","useCases":["Let users sign up and log in","Connect Clerk or Auth0 for sign-in","Use a built-in login library"],"pricing":"Included.","status":"ga","url":"https://docs.convex.dev/auth","source":"https://www.convex.dev/pricing"},{"id":"convex-search","provider":"convex","name":"Text and Vector Search","category":"vector","also":["ai-search"],"tagline":"Search your app's text and meaning, right inside your database.","description":"Text and Vector Search lets your app find things by keyword or by meaning, using the same database that holds your data. If you are building a site search box or a chatbot that answers from your own documents (often called RAG, where an AI looks up your content before replying), this covers both. Vector search means matching by meaning rather than exact words. You skip adding a separate search service. Search storage is billed separately, at $0.55 per GB on the Starter plan.","useCases":["Search text across app content","Find similar items using vectors","Power retrieval for AI answers"],"pricing":"$0.55/GB search storage on Starter.","status":"ga","url":"https://docs.convex.dev/search","source":"https://www.convex.dev/pricing"},{"id":"convex-crons","provider":"convex","name":"Scheduled Functions and Crons","category":"workflows","tagline":"Run tasks later or on a schedule, like reminders and nightly jobs.","description":"Scheduled Functions and Crons let your app run a piece of code at a later time or on a repeating schedule. Crons are jobs that repeat, such as every night at midnight. If your app sends reminders, you can schedule each one the moment a user sets it. Jobs are scheduled transactionally, meaning the schedule is saved together with your other data changes, so the two stay in step. They are included, though each run counts as a function call.","useCases":["Send reminders on a schedule","Run jobs at set times","Run timed jobs with safe updates"],"pricing":"Counts as function calls.","status":"ga","url":"https://docs.convex.dev/scheduling","source":"https://www.convex.dev/pricing"},{"id":"convex-components","provider":"convex","name":"Convex Components","category":"devtools","also":["agents"],"tagline":"Ready-made backend building blocks, like an AI agent or rate limiter.","description":"Convex Components are reusable pieces of backend that you add to your project instead of writing yourself. Official ones include Agent, Rate Limiter (which stops anyone from making too many requests) and Workflow. If you want your AI assistant to remember past conversations, you can add the Agent component with npm, the standard tool for installing code packages. They are free. Pick these when you want a ready-made building block rather than building and maintaining your own.","useCases":["Add AI agent memory","Install ready-made backend modules","Add rate limits to your app"],"pricing":"Free.","status":"ga","url":"https://www.convex.dev/components","source":"https://www.convex.dev/pricing"},{"id":"convex-chef","provider":"convex","name":"Chef","category":"ai-apps","tagline":"An AI app builder from Convex for trying out full-stack app ideas.","description":"Chef is an AI app builder made by Convex. If you have an idea and want a quick prototype of a full-stack app, meaning one with both the screens people see and the data behind them, it is a handy place to start. You can use it to test an idea before committing to a bigger build. Because it comes from the Convex team, it is a natural pick if you already plan to use Convex. Pricing has not been confirmed, so check before you plan around it.","useCases":["Prototype full-stack apps quickly","Generate a starter app with AI","Test an app idea before building"],"pricing":"Not verified.","status":"ga","url":"https://chef.convex.dev/","source":"https://www.convex.dev/pricing"},{"id":"convex-dedicated","provider":"convex","name":"Dedicated Deployments","category":"hosting","tagline":"Your own private Convex setup for apps with very heavy traffic.","description":"Dedicated Deployments give you a Convex setup used only by your app, rather than shared with other customers. This is called single-tenant. If your product has grown to heavy traffic and you want more room to scale, this is the step up. It is available on the Business and Enterprise plans. Expect a $2,500 monthly minimum, so it makes sense for established products, not early experiments.","useCases":["Run very large apps on dedicated capacity","Keep one customer's backend isolated","Expect a $2,500 monthly minimum"],"pricing":"$2,500 monthly minimum.","status":"ga","url":"https://www.convex.dev/pricing","source":"https://www.convex.dev/pricing"},{"id":"mongodb-atlas","provider":"mongodb","name":"MongoDB Atlas","category":"nosql","tagline":"A hosted MongoDB database that runs on AWS, Azure or Google Cloud.","description":"MongoDB Atlas is a managed version of MongoDB, a database that stores information as flexible documents instead of rigid tables. Managed means MongoDB runs it for you on AWS, Azure or Google Cloud. If you are building an app whose data changes shape often, like user profiles with different fields, documents are easy to work with. There is a free M0 tier to start, and paid plans are priced by cluster tier, meaning the size of the database setup you choose. MongoDB 9.0 is the latest generally available release.","useCases":["Run a managed MongoDB database","Host data on AWS, Azure or Google","Store app data as documents"],"pricing":"Free M0; paid by cluster tier.","status":"ga","url":"https://www.mongodb.com/products/platform/atlas-database","source":"https://www.mongodb.com/products"},{"id":"mongodb-atlas-search","provider":"mongodb","name":"Atlas Search","category":"ai-search","tagline":"Add a search box to your app using the data you already keep in Atlas.","description":"Atlas Search is full-text search built into MongoDB Atlas, so people can type words and find matching content. If you run a store or blog and want a site search box, you can search your data right where it lives. It is based on Lucene, a widely used open source search engine, and sits next to your data instead of in a separate tool. That means no second service to keep in sync. It is included with your cluster.","useCases":["Add full-text search to your site","Search text in the same database","Search text next to your data"],"pricing":"Included with cluster.","status":"ga","url":"https://www.mongodb.com/products/platform/atlas-search","source":"https://www.mongodb.com/products"},{"id":"mongodb-atlas-vector-search","provider":"mongodb","name":"Atlas Vector Search","category":"vector","tagline":"Find things by meaning, not just keywords, inside MongoDB Atlas.","description":"Atlas Vector Search lets you store embeddings, which are lists of numbers that capture the meaning of text or images, and then search through them. If you are building a chatbot that answers from your own documents (known as RAG), this helps find the most relevant passages to hand to the AI. Because it lives inside Atlas, your embeddings can sit alongside the rest of your data. It is included with your cluster.","useCases":["Store embeddings for AI features","Find similar items by meaning","Feed retrieved context into AI apps"],"pricing":"Included with cluster.","status":"ga","url":"https://www.mongodb.com/products/platform/atlas-vector-search","source":"https://www.mongodb.com/products"},{"id":"mongodb-voyage","provider":"mongodb","name":"Voyage AI","category":"ai-models","tagline":"AI models that help your app find the most relevant information.","description":"Voyage AI makes embedding and reranking models. Embedding models turn text into numbers that capture meaning, and reranking models take a list of search results and put the best ones first. If your chatbot sometimes pulls up the wrong documents, better retrieval models can improve its answers. MongoDB acquired Voyage AI in 2025, so it is now part of the MongoDB family. Pricing is per token, which is a small chunk of text, and there is a free allowance.","useCases":["Create embeddings for search and AI","Rerank results to improve relevance","Use a free token allowance first"],"pricing":"Per token; free allowance.","status":"ga","url":"https://www.mongodb.com/products/platform/ai-search-and-retrieval","source":"https://www.mongodb.com/products"},{"id":"mongodb-stream-processing","provider":"mongodb","name":"Atlas Stream Processing","category":"data-pipelines","tagline":"React to a live flow of events as they happen, using MongoDB queries.","description":"Atlas Stream Processing works on data continuously as it arrives, rather than waiting to handle it in batches. If your app tracks events like orders, clicks or sensor readings and needs to react right away, you can filter and reshape them as they flow in. It uses MongoDB aggregation syntax, the same query style used for MongoDB data, so there is less new to learn if you already know it. Pricing is by hourly processor tiers.","useCases":["Process event data as it arrives","Write pipelines in familiar MongoDB syntax","Handle streams for event-driven apps"],"pricing":"Hourly processor tiers.","status":"ga","url":"https://www.mongodb.com/products/platform/atlas-stream-processing","source":"https://www.mongodb.com/products"},{"id":"mongodb-data-federation","provider":"mongodb","name":"Atlas Data Federation","category":"warehouse","tagline":"Search across your database and file storage with one question.","description":"Atlas Data Federation lets you ask one question across data that lives in different places, such as your MongoDB Atlas database and files in Amazon S3, a popular file storage service. If your app has years of old orders piling up, you can move them out with Online Archive and still search them when you need to. It is a good pick when you want reports across old and new data without copying everything into one place first. You pay per terabyte of data your queries process.","useCases":["Query Atlas, S3 and other sources","Search old data in archives","Pay per terabyte processed"],"pricing":"Per TB processed.","status":"ga","url":"https://www.mongodb.com/products/platform/atlas-data-federation","source":"https://www.mongodb.com/products"},{"id":"mongodb-charts","provider":"mongodb","name":"Atlas Charts","category":"warehouse","tagline":"Build charts and dashboards from your MongoDB data, free.","description":"Atlas Charts turns the data in your MongoDB Atlas database into charts and dashboards you can look at and share. If you run a small app and want to see signups, orders or usage without exporting spreadsheets, you can build a dashboard right on top of your data. Pick it over a separate reporting tool when your data already lives in Atlas and you want to skip extra setup. It is free.","useCases":["Build dashboards from MongoDB data","Make charts for team reports","Turn database records into charts"],"pricing":"Free.","status":"ga","url":"https://www.mongodb.com/products/platform/atlas-charts","source":"https://www.mongodb.com/products"},{"id":"mongodb-triggers","provider":"mongodb","name":"Atlas Triggers","category":"workflows","tagline":"Run your code automatically when data changes or on a schedule.","description":"Atlas Triggers runs a small piece of your code automatically, either when something changes in your database or on a schedule. If you want to send a welcome email when a new user is saved, or clean up old records every night, a trigger does it without you running anything by hand. Pick it when the action depends on what happens in your MongoDB data. It remains available after the wider App Services product reached end of life on September 30, 2025, and it is included with Atlas.","useCases":["Run code when data changes","Run tasks on a schedule","React to database updates automatically"],"pricing":"Included.","status":"ga","url":"https://www.mongodb.com/docs/atlas/atlas-ui/triggers/","source":"https://www.mongodb.com/products"},{"id":"mongodb-agent-engine","provider":"mongodb","name":"Atlas Agent Engine","category":"agents","tagline":"Build and manage AI agents that remember and look things up.","description":"Atlas Agent Engine is a place to build, run and keep control of AI agents, which are programs that use an AI model to complete tasks for you. If you are making an app where an assistant needs to remember past conversations and look up your own information, this gives you memory and retrieval, meaning the lookup of relevant facts, in one spot. Pick it when you already use MongoDB Atlas and want your agent tools from the same provider. It is in public preview, and pricing has not been published.","useCases":["Build AI agents with memory","Add retrieval to agent workflows","Test agents in public preview"],"pricing":"Not published.","status":"preview","url":"https://www.mongodb.com/products","source":"https://www.mongodb.com/products"},{"id":"mongodb-infinite","provider":"mongodb","name":"Atlas Infinite","category":"nosql","tagline":"A MongoDB database whose power and storage scale separately.","description":"Atlas Infinite is a version of the MongoDB Atlas database where the computing power and the storage space grow and shrink independently of each other. If your app gets sudden bursts of traffic, such as a product launch or a viral post, you can add power for the rush without also adding storage you do not need. Pick it over a standard setup when your usage jumps around instead of staying steady. It is in public preview, and pricing has not been published.","useCases":["Handle spiky traffic workloads","Scale compute and storage separately","Try in public preview"],"pricing":"Not published.","status":"preview","url":"https://www.mongodb.com/products","source":"https://www.mongodb.com/products"},{"id":"mongodb-enterprise-advanced","provider":"mongodb","name":"MongoDB Enterprise Advanced","category":"nosql","tagline":"MongoDB you run on your own servers, with enterprise tools.","description":"MongoDB Enterprise Advanced is the paid edition of MongoDB that you run yourself, on your own servers or in a private cloud, with extra tooling aimed at larger companies. If you work in a regulated field like healthcare or banking and your data must stay on machines you control, this lets you keep it there. Pick it over Atlas, MongoDB's hosted service, when rules or company policy say you cannot use a hosted database. Pricing is available by contacting sales.","useCases":["Self-host MongoDB on your servers","Run in private cloud","Host regulated data on your own"],"pricing":"Contact sales.","status":"ga","url":"https://www.mongodb.com/products/self-managed/enterprise-advanced","source":"https://www.mongodb.com/products"},{"id":"mongodb-community","provider":"mongodb","name":"MongoDB Community Edition","category":"nosql","tagline":"The free version of MongoDB you run yourself, great for local testing.","description":"MongoDB Community Edition is the free version of MongoDB that you download and run yourself. If you want to try an idea on your laptop before paying for anything, you can install it and build against a real MongoDB database. Pick it over Atlas, the hosted version, when you just want to experiment or develop locally. Know that it comes with no support and no commercial license, so you are on your own if something breaks. Its code is source-available, meaning you can read it, though usage rules apply.","useCases":["Run MongoDB on your laptop","Develop locally without paying","Try MongoDB before buying support"],"pricing":"Free.","status":"ga","url":"https://www.mongodb.com/products/self-managed/community-edition","source":"https://www.mongodb.com/products"},{"id":"mongodb-compass","provider":"mongodb","name":"MongoDB Compass","category":"devtools","tagline":"A free desktop app to browse and search your MongoDB data.","description":"MongoDB Compass is a free desktop app that lets you look at your MongoDB data and search it by clicking instead of writing code. If you built an app with an AI tool and want to check whether new users are really being saved, you can open Compass, connect to your database and see the records for yourself. Pick it when you want to inspect your data and a blank command window feels intimidating. It is free.","useCases":["Browse and query database visually","Inspect records in a desktop app","Explore data without writing code"],"pricing":"Free.","status":"ga","url":"https://www.mongodb.com/products/tools/compass","source":"https://www.mongodb.com/products"},{"id":"mongodb-atlas-cli","provider":"mongodb","name":"Atlas CLI","category":"devtools","tagline":"Set up and manage Atlas databases by typing commands, free.","description":"Atlas CLI is a free command-line tool, which means you use it from the terminal, the text window where you type commands, to manage your MongoDB Atlas projects and databases. If you often set up the same kind of database for new projects, you can write the steps once and repeat them instead of clicking through the website each time. It also runs local Atlas environments, so you can test on your own computer. Pick it over the web dashboard when you want setup to be quick and repeatable.","useCases":["Create and manage Atlas from terminal","Set up local Atlas environments","Automate database setup steps"],"pricing":"Free.","status":"ga","url":"https://www.mongodb.com/docs/atlas/cli/","source":"https://www.mongodb.com/products"},{"id":"mongodb-mcp-server","provider":"mongodb","name":"MongoDB MCP Server","category":"agents","tagline":"Let your AI tools read and work with your MongoDB data, free.","description":"MongoDB MCP Server is the official bridge that lets AI tools talk to your MongoDB data. MCP, short for Model Context Protocol, is a common way for AI assistants to connect to outside services. If you use an AI coding assistant and want it to look at your real data while you build, this lets it do that directly. Pick it when you want your AI tool to work with your database instead of you pasting in examples by hand. It is free.","useCases":["Let AI tools read MongoDB data","Connect AI assistants to databases","Give agents access to data"],"pricing":"Free.","status":"ga","url":"https://www.mongodb.com/products","source":"https://www.mongodb.com/products"},{"id":"mongodb-relational-migrator","provider":"mongodb","name":"Relational Migrator","category":"migration","tagline":"Move your data from a SQL database into MongoDB, free.","description":"Relational Migrator helps you move data from a traditional database, the kind that stores information in linked tables and is queried with SQL, into MongoDB. If your app started on one of those and you want to switch, it helps map how your tables should look in MongoDB and then moves the data over. Pick it over doing the move by hand when your data has many connected tables and you want the structure planned for you. It is free.","useCases":["Move data from SQL databases","Map old schemas to MongoDB","Migrate tables into documents"],"pricing":"Free.","status":"ga","url":"https://www.mongodb.com/products/tools/relational-migrator","source":"https://www.mongodb.com/products"},{"id":"mongodb-device-sync","provider":"mongodb","name":"Atlas Device Sync and Data API","category":"realtime-db","tagline":"Mobile sync and web data access from MongoDB, now retired.","description":"Atlas Device Sync and Data API offered two things: keeping the data in a mobile app in step with the cloud database, and a REST Data API, a simple way for apps to read and write data over the web. If you maintained an older mobile app built around this sync, it is what kept phones and the database matching. It is not a choice for anything new. It reached end of life on September 30, 2025, along with App Services hosting and GraphQL, so plan to move off it.","useCases":["Sync data for old mobile apps","Retired September 30, 2025","Not available for new apps"],"pricing":"Retired.","status":"closed","url":"https://www.mongodb.com/docs/atlas/app-services/deprecation/","source":"https://www.mongodb.com/products"},{"id":"cockroachdb-continuum-standard","provider":"cockroachdb","name":"Cockroach Continuum Standard","category":"sql","tagline":"A hosted database that grows with your app and bills by usage.","description":"Cockroach Continuum Standard is a hosted database where your app keeps its data in tables you can search with SQL, the common database language. If you are building a SaaS product (software people subscribe to), it is a sensible home for accounts, orders and settings. It runs on shared infrastructure, meaning you share servers with other customers, and it carries a 99.99% uptime promise. Storage is billed by use, and compute starts at $0.092 per vCPU-hour (a unit of processor time) on AWS us-east-1. It is the edition for new Cloud organizations since September 15, 2026.","useCases":["Run the database for a SaaS product","Pay for storage and compute as used","Keep the database running 99.99% of the time"],"pricing":"$0.092/vCPU-hour; trial $400.","status":"ga","url":"https://www.cockroachlabs.com/pricing/","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-continuum-mission-critical","provider":"cockroachdb","name":"Cockroach Continuum Mission Critical","category":"sql","tagline":"A dedicated database for apps that cannot go down, spread across regions.","description":"Cockroach Continuum Mission Critical is a CockroachDB database that runs on its own dedicated servers instead of shared ones, built for data you cannot afford to lose or be locked out of. If your product handles card payments or health records, it adds the compliance options you need: PCI (card payment rules), HIPAA (US health privacy rules) and CMEK, which lets you control the encryption keys protecting your data. It promises 99.999% uptime across multiple regions and includes all add-ons. Pick it over Standard when downtime would be a crisis; compute is $0.162 per vCPU-hour.","useCases":["Run business-critical systems across multiple regions","Meet PCI and HIPAA requirements","Host virtual clusters for many customers"],"pricing":"$0.162/vCPU-hour.","status":"ga","url":"https://www.cockroachlabs.com/pricing/","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-legacy-cloud-plans","provider":"cockroachdb","name":"CockroachDB Cloud Basic, Standard and Advanced","category":"sql","tagline":"Older CockroachDB Cloud plan names, kept for customers who already use them.","description":"CockroachDB Cloud Basic, Standard and Advanced are the earlier plan names for CockroachDB's hosted database. Basic was previously called Serverless, and Advanced was called Dedicated. If you already run a cluster (a group of servers working together as one database) on one of these, it stays on that plan unless you choose to move to Continuum. If you are starting something new, look at the Continuum plans instead. These plans are closed to new customers.","useCases":["Keep existing clusters on current plans","Keep current pricing until opting in","Move existing clusters to Continuum later"],"pricing":"Existing pricing.","status":"closed","url":"https://docs.cockroachlabs.com/docs/releases/cloud","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-self-hosted","provider":"cockroachdb","name":"CockroachDB Self-Hosted","category":"sql","also":["vms"],"tagline":"Run CockroachDB on your own servers, in your own data center or cloud.","description":"CockroachDB Self-Hosted is the version of the database you install and run yourself, rather than having it hosted for you. If your company must keep data in its own data center or inside a specific country's rules, this lets you do that on any infrastructure you choose. The tradeoff is that you handle setup, upgrades and monitoring yourself, which is more work than the hosted plans. It needs an enterprise license, and the free core edition ended in 2024, so contact sales for pricing.","useCases":["Run the database in your own data center","Keep data in a sovereign cloud","Run it on any infrastructure you choose"],"pricing":"License; contact sales.","status":"ga","url":"https://www.cockroachlabs.com/product/","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-cdc","provider":"cockroachdb","name":"Change Data Capture (CDC)","category":"data-pipelines","tagline":"Send a live feed of database changes to other tools as they happen.","description":"Change Data Capture, or CDC, watches your database and sends out a message each time a row is added, changed or removed. If you want your app to email a customer when an order updates, or keep a search tool or analytics system in sync, you can feed those changes to Kafka (a popular event-streaming tool) and other destinations. It saves you from repeatedly asking the database what is new. It is an add-on billed per vCPU-hour (a unit of processor time) only while active, at $0.006 on Standard.","useCases":["Stream database changes to Kafka","Feed event updates to other apps","Send changes to other destinations"],"pricing":"$0.006/vCPU-hr Standard.","status":"ga","url":"https://docs.cockroachlabs.com/docs/continuum/editions-and-add-ons","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-pcr","provider":"cockroachdb","name":"Physical Cluster Replication","category":"sql","tagline":"Keep a live backup copy of your database in another region for failover.","description":"Physical Cluster Replication keeps a continuous copy of your database in a second cluster, in another region. If your main region has an outage, you can switch over to the copy and your app keeps running with little lost data. It is aimed at disaster recovery, meaning planning for the worst day, rather than everyday backups. It is a limited access add-on in preview, so it is still early, and it costs $0.023 per vCPU-hour (a unit of processor time).","useCases":["Keep a live copy in another region","Fail over if a region goes down","Recover quickly from regional outages"],"pricing":"$0.023/vCPU-hour.","status":"preview","url":"https://docs.cockroachlabs.com/docs/continuum/editions-and-add-ons","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-virtual-clusters","provider":"cockroachdb","name":"Cluster Virtualization","category":"sql","tagline":"Give each customer their own isolated database inside one shared cluster.","description":"Cluster Virtualization lets you split one large CockroachDB setup, called the host cluster, into several isolated virtual clusters, each acting like its own separate database. If you run a SaaS product where each business customer needs its data kept apart from everyone else's, this gives each one a walled-off space without you managing many separate databases. It is a preview add-on on Continuum, so it is still early, and it costs $0.023 per vCPU-hour (a unit of processor time).","useCases":["Give each customer an isolated database","Host many customers on one cluster","Separate customer data inside shared infrastructure"],"pricing":"$0.023/vCPU-hour.","status":"preview","url":"https://docs.cockroachlabs.com/docs/continuum/overview","source":"https://www.cockroachlabs.com/pricing/"},{"id":"cockroachdb-molt","provider":"cockroachdb","name":"MOLT","category":"migration","tagline":"Free tools for moving your data from Postgres or MySQL to CockroachDB.","description":"MOLT is a free toolkit for moving an existing database over to CockroachDB. It handles three jobs: converting your schema (the layout of your tables), loading your data, and verifying that everything arrived correctly. If your app currently runs on Postgres or MySQL, two popular databases, and you want to switch, it saves you from piecing together your own scripts. Use it when you want a checked, step-by-step move instead of a risky one-time copy.","useCases":["Move a Postgres database over","Move a MySQL database over","Check copied data matches the source"],"pricing":"Free tooling.","status":"ga","url":"https://docs.cockroachlabs.com/docs/molt/migration-overview","source":"https://www.cockroachlabs.com/pricing/"}]