All vikis
007 First Light
AI
Aion 2
Ananta
ARC Raiders
ArcheAge Chronicles
Call of Duty: Modern Warfare 4
ChatGPT
Chrono Odyssey
Civilization 3
Civilization 4
Crimson Desert
Deadlock
Diablo 4
Diablo 5
Enshrouded
Fable
Final Fantasy Resonance
Football Manager 27
Forza Horizon 6
God of War
Grand Theft Auto 6
Heroes of Might and Magic 3Claude on Cloud Platforms
Besides Anthropic's own Claude API, Claude models are sold through Amazon Bedrock, Claude Platform on AWS, Google Cloud (Google managed or Anthropic managed) and Microsoft Foundry. The Anthropic-operated options get new API features the same day, while Bedrock and Google managed run Claude on the cloud provider's own infrastructure with fewer features.
Claude models are available in five places besides Anthropic's own Claude API: Amazon Bedrock and Claude Models on Google Cloud (Google managed), the former "Claude on Vertex AI", are run by the cloud providers; Claude Platform on AWS, Claude Models on Google Cloud (Anthropic managed) and Microsoft Foundry are run by Anthropic and billed through the cloud's marketplace (API overview, as of Nov 6, 2026). The choice decides which features you get, who processes your data and how you are billed. Token prices on the Anthropic-operated options follow Anthropic's list prices, while Bedrock and Google bill at their own published prices.
Quick facts
| Operated by | Billing | Request size | Managed Agents | |
|---|---|---|---|---|
| Claude API | Anthropic | Anthropic, USD | 128 MB | Yes (beta) |
| Claude Platform on AWS | Anthropic | AWS Marketplace, CCUs | 32 MB | Yes (beta) |
| Google Cloud (Anthropic managed) | Anthropic | Google Cloud Marketplace, CCUs | 32 MB | Yes (beta) |
| Microsoft Foundry | Anthropic, hosted on Azure or on Anthropic | Azure Marketplace, CCUs | 32 MB | No |
| Amazon Bedrock | AWS | AWS, at AWS prices | 20 MB | No |
| Google Cloud (Google managed) | Google Cloud, at Google prices | 30 MB | No |
Source: API overview and Pricing, Anthropic docs, read Nov 6, 2026. A CCU (Claude Consumption Unit) costs a fixed $0.01.
Amazon Bedrock
Bedrock runs Claude on AWS's own infrastructure. For Claude Opus 4.7 and later it serves the Messages API at /anthropic/v1/messages, with Anthropic staff having no access to the inference machines; older models use Bedrock's InvokeModel and Converse APIs with IDs such as anthropic.claude-opus-4-6-v1 (Claude in Amazon Bedrock). Some models are open to every Bedrock customer, while Opus 5.5, Opus 5, Sonnet 5.5 and Haiku 5.5 have their own access criteria in the AWS console.
What Bedrock lacks, per Anthropic's docs: the Files API, server tools (the Code Execution Tool, web search, web fetch, advisor), Agent Skills, the MCP connector, programmatic tool calling, the Message Batches API, the Models and Admin APIs, and Managed Agents. It has the Messages API, Prompt Caching, thinking, client tools including Computer Use, citations and structured outputs. Its default quota is 2 million input tokens per minute, raisable to 5 million input and 500,000 output without Anthropic's approval. Regional endpoints cost 10% more than global ones for Claude 4.5 models and later.
Anthropic's docs point to Bedrock for FedRAMP High, IL4, IL5 or HIPAA needs, or when AWS must be the sole data processor (Claude Platform on AWS). Bedrock also keeps some models after Anthropic retires them, such as Opus 4.1.
Claude Platform on AWS
Launched May 11, 2026, this is the Claude API itself, run by Anthropic but reached with AWS credentials (IAM SigV4 or an API key), PrivateLink and AWS Marketplace billing (Claude Platform on AWS). It uses the same paths and model IDs as the Claude API, and new features arrive "typically same-day". It has Agent Skills, code execution, batches, the Files API, customer-managed keys in AWS KMS and Managed Agents. It lacks HIPAA readiness, most of the Admin API, fast mode and the self-service tier increase. Billing is postpaid in CCUs, and Max, Ultra or Team plan API credits do not apply.
Google Cloud: Google managed or Anthropic managed
Google managed is Claude on Google's Vertex AI, which Anthropic's docs now describe under Google's "Agent Platform" name (API aiplatform.googleapis.com). The model goes in the URL, the version in the body, and IDs before Claude 4.6 use @ (claude-haiku-4-5@20251001). Google sets rate limits, billing and data handling. It lacks code execution, web fetch, the Files API, batches, Agent Skills and Managed Agents, and has only basic web search (Google managed).
Anthropic managed, launched Oct 22, 2026, is the Claude API run by Anthropic and reached through a Google Cloud project, with Google Cloud IAM for access and CCUs on the Google Cloud bill that count toward Google spend commitments (Anthropic managed). It has Managed Agents, fast mode for Opus 5.5 and message threads, but no API keys (Google credentials only) and no MCP tunnels.
Microsoft Foundry
Claude has been in Microsoft Foundry since Nov 18, 2025, billed through the Azure Marketplace with Azure or Entra ID sign-in. There are two hosting options (Claude in Microsoft Foundry):
- Hosted on Azure: Anthropic's service on Azure infrastructure for the latest Opus, Sonnet and Haiku, where prompts and outputs stay in Azure and only usage data and flagged content go to Anthropic. It lacks code execution, Agent Skills, the Files API and the newer web tools, and offers a US Data Zone deployment. Fast mode for Opus 5.5 arrived Oct 15, 2026.
- Hosted on Anthropic: every Claude model, with more API features.
Neither option has the Admin API, batches, the advisor tool or Managed Agents. HIPAA is possible with a BAA plus Azure configuration.
Which should you choose?
| If you need | Choose |
|---|---|
| Every feature first, simplest setup | Claude API |
| AWS billing and IAM, with Claude API features | Claude Platform on AWS |
| FedRAMP High, IL4/IL5, or AWS as sole processor | Amazon Bedrock |
| Google Cloud billing and commitments, with Claude API features | Google Cloud (Anthropic managed) |
| Google to run and govern the service | Google Cloud (Google managed) |
| Azure billing, Entra ID, data kept in Azure | Microsoft Foundry, hosted on Azure |
The trade-off is plain: the cloud-run options give procurement and compliance benefits, while the Anthropic-run ones get features sooner. Claude Platform on AWS also draws on a capacity pool separate from both the Claude API and Bedrock (Claude Platform on AWS). For data handling across platforms, see Claude Privacy and Data Use.
Frequently asked questions
Is Claude cheaper on Amazon Bedrock than on the Claude API? Not by Anthropic's list: Bedrock bills Claude at AWS's own published prices, which you should compare on AWS's pricing page, and regional Bedrock endpoints cost 10% more than global ones for Claude 4.5 models and later (Pricing).
Is Claude still on Vertex AI? Yes. Claude on Vertex AI is now called Claude Models on Google Cloud (Google managed) in Anthropic's docs, and since Oct 22, 2026 there is also an Anthropic managed option on Google Cloud.
Does Claude Code work with Bedrock, Vertex AI or Foundry? Yes. The Claude Code terminal CLI, VS Code and JetBrains extensions can run against Amazon Bedrock, Google Cloud's Agent Platform (Vertex AI) or Microsoft Foundry instead of an Anthropic account (Claude Code overview); see Claude Code for setup.
Can I use my Claude Max API credits on Bedrock? No. The monthly API credits in Claude Max, Ultra and Team plans work only on the Claude API in the Console, not on any cloud platform (API credits).
Which cloud platforms have Claude Managed Agents? Claude Managed Agents runs only on the Claude API, Claude Platform on AWS and Claude Models on Google Cloud (Anthropic managed), as of Nov 6, 2026.
Sources
- API overview, Anthropic docs, read Nov 6, 2026
- Pricing, Anthropic docs, read Nov 6, 2026
- Claude in Amazon Bedrock, Anthropic docs, read Nov 6, 2026
- Claude Platform on AWS, Anthropic docs, read Nov 6, 2026
- Claude Models on Google Cloud (Google managed), Anthropic docs, read Nov 6, 2026
- Claude Models on Google Cloud (Anthropic managed), Anthropic docs, read Nov 6, 2026
- Claude in Microsoft Foundry, Anthropic docs, read Nov 6, 2026
- API credits for Max, Ultra, and Team plans, Anthropic docs, read Nov 6, 2026
- Release notes, Anthropic docs, read Nov 6, 2026
- Claude Code overview, Anthropic, read Nov 6, 2026
Related
- Claude API PricingClaude API prices are per million tokens, from $0.10 input and $0.50 output on Claude Haiku 5.5 to $10 and $50 on Claude Fable 5.1, with Opus 5.5 at $4 and $20 and Sonnet 5.5 at $2 and $10, as of Nov 6, 2026. Prompt caching and the 50% Batch discount lower that.
- Claude API Rate LimitsClaude API rate limits are set per organization by usage tier (Start, Build, Scale, Custom) as requests, input tokens and output tokens per minute; on the Start tier Claude Opus 5.5 allows 4,000 requests, 2,000,000 input tokens and 400,000 output tokens a minute. Going over returns a 429 error, while a 529 error means the API itself is overloaded.
- Claude API SDKsAnthropic publishes official Claude API SDKs in seven languages (Python, TypeScript, Java, Go, Ruby, C# and PHP) plus the ant command-line tool; the Python SDK installs with pip install anthropic and the TypeScript SDK with npm install @anthropic-ai/sdk. Since Oct 21, 2026 the new major versions are Apache 2.0 licensed.
-
Claude ConsoleThe Claude Console at platform.claude.com is Anthropic's web dashboard for the Claude API, where developers create API keys, test prompts in the playground, organize workspaces, watch usage and rate limits, and buy credits. It was called the Anthropic Console until Sep 2025, and its Workbench became the playground in Aug 2026.
-
Computer UseComputer use is a Claude API tool that lets Claude operate a desktop through screenshots, mouse clicks and keystrokes, which your own application carries out. It launched in public beta on Oct 22, 2024 and has been generally available as the computer_toolset_20260801 toolset since Aug 19, 2026.
-
Messages APIThe Messages API is the Claude API's main endpoint, POST /v1/messages: you send a model, max_tokens and a list of user and assistant messages, and Claude returns a message with content blocks, a stop_reason and token usage, whole or streamed.
-
Model Context ProtocolThe Model Context Protocol (MCP) is an open standard, introduced by Anthropic on Nov 25, 2024, for connecting AI applications such as Claude to outside tools and data through MCP servers. OpenAI, Google and Microsoft support it, and Anthropic donated it to the Linux Foundation's Agentic AI Foundation on Dec 9, 2025.
- Structured OutputsStructured outputs make the Claude API return JSON that always matches a JSON Schema you supply, through output_config.format, or tool inputs that match exactly, through strict tool use. It is Claude's answer to a JSON mode, generally available since Jan 29, 2026.
See also
-
Claude APIThe Claude API is Anthropic's REST API at api.anthropic.com for building with Claude models: create an API key in the Claude Console at platform.claude.com, then send requests to the Messages API, paying per token.
- Files APIThe Claude Files API stores a PDF, text file, image or dataset once with POST /v1/files and returns a file_id that later Claude API requests reference instead of re-uploading. Files up to 500 MB, 1 TB per organization, free to store; out of beta since Aug 19, 2026.
-
Code Execution ToolThe Claude API's code execution tool runs Bash commands and Python in a sandboxed Linux container at Anthropic, with 5 GiB of memory and no internet access. Each organization gets 1,550 free container hours a month, then $0.05 per hour, and it is free when used with the newer web search or web fetch tools.
- Message Batches APIThe Message Batches API runs up to 100,000 Claude API requests asynchronously at 50% off the normal price; most batches finish within an hour, every batch within 24 hours, and results stay downloadable for 29 days.
-
-
Prompt CachingPrompt caching in the Claude API stores a prompt's prefix so later requests reuse it: you add cache_control, the first request writes the cache at 1.25x the input price, and reads within 5 minutes (or 1 hour) cost 0.1x or less.