Debuly
Claude Platform logo

Claude Platform

Anthropic's developer platform and API for building apps and agents on Claude models, with per-token pricing by model.

Paid
Claude Platform preview image from its official website
Image from platform.claude.com · Vendor marketing image from the official site; removed on request

What is Claude Platform?

Claude Platform is where developers build with Claude through Anthropic's API. The docs organize the API into five areas: model capabilities such as reasoning effort and citations, tools that let Claude act on the web or in your environment, tool infrastructure, context management for long sessions, and files and assets. Alongside the Messages API, which gives direct prompting control for custom agent loops, there is Claude Managed Agents for long-running work on managed infrastructure. Administration and governance are covered by an Admin API, a Usage and Cost API and a Compliance API. Models come in four families (Fable, Opus, Sonnet and Haiku), and the API is also offered through Amazon Bedrock, Claude Platform on AWS, Google Cloud and Microsoft Foundry. Pricing is per million tokens. Claude Haiku 5.5 starts at $0.10 per million input tokens, Sonnet 5.5 is $2 input and $10 output, and batch calls cost 50% less than standard calls.

Use cases

  • Add Claude to your own application
  • Build tool-using AI agents
  • Process large request volumes in batches
  • Ground answers in documents with citations
  • Cut repeat costs with prompt caching
  • Monitor team usage and spend

Who it's for

  • Developers
  • AI engineers
  • Enterprise platform teams

Why use Claude Platform

  • Per-token billing means you pay for what an application actually uses, and a model ladder from Haiku to Fable lets you match cost to task difficulty.
  • Batch processing at 50% less than standard calls helps with large offline jobs such as classification or extraction.
  • Prompt caching rates make repeated long prompts cheaper than sending them fresh each time.
  • The same models are reachable through Bedrock, Google Cloud and Microsoft Foundry if your company buys cloud services from one of those vendors.
  • Separate Usage and Cost and Admin APIs make it possible to track spend programmatically.

Claude Platform features

  • Four model families: Fable, Opus, Sonnet and Haiku
  • Messages API for direct prompting and custom agent loops
  • Tools that let Claude act on the web or your environment
  • Prompt caching with separate cache write and hit rates
  • Batch processing at 50% less than standard API calls
  • Citations that point to exact passages in source documents
  • Adjustable thinking depth through the effort parameter
  • Context management features for long-running sessions
  • Admin API, Usage and Cost API and Compliance API
  • Claude Managed Agents for long-running, asynchronous tasks
  • Zero Data Retention eligibility marked per feature

Claude Platform integrations

  • Amazon Bedrock
  • Google Cloud
  • Microsoft Foundry
  • AWS

How to get started with Claude Platform

  1. Create an account and an API key, setting an expiration and keeping the key out of source control and client-side code.
  2. Install an SDK and send a first message through the Messages API to see the request and response shape.
  3. Choose a model by comparing capability and cost; the intro suggests Claude Opus 5.5 as a starting point for most workloads.
  4. Add tools, citations, caching or batch processing as your application needs them.
  5. Monitor usage and cost with the Usage and Cost API and organization admin controls.

Claude Platform limitations

  • Feature availability differs between the Claude API and cloud partner platforms
  • Some features are in beta and may change or be discontinued
  • Batch processing is not Zero Data Retention eligible
  • Older models such as Claude Opus 4 and 4.1 are retired outside some cloud providers

Claude Platform pricing

Claude Platform has no free plan. Plans start at $0.1 usage-based, according to the official pricing page on .

Claude Platform plans

  • Claude Haiku 5.5

    $0.10 /MTok input

    Output $0.50 / MTok, prompts up to 100,000 tokens

    • Fastest, lowest-cost family
    • High-volume, latency-sensitive tasks
  • Claude Sonnet 5.5

    $2 /MTok input

    Output $10 / MTok

    • Everyday tasks and writing
    • Cost-efficient
  • Claude Opus 5.5

    $4 /MTok input

    Output $20 / MTok

    • Long-running agentic coding
    • Knowledge work
  • Claude Fable 5.1

    $10 /MTok input

    Output $50 / MTok

    • Demanding reasoning
    • Long-horizon agentic work

Plans as listed on the official pricing page on . Prices can change and may vary by country or exclude tax. Check the official pricing page.

Claude Platform alternatives

The main Claude Platform alternatives listed on Debuly are OpenAI API, Gemini API, Mistral AI, Cohere, OpenLIT and OnitsAI. The table compares pricing and platforms; order is A–Z.

Claude Platform compared with its alternatives
ToolPricingStarting pricePlatforms
Claude PlatformPaid$0.1, usage-basedAPI, Web
CohereFreemium—API, Web, Private deployment
Gemini APIFreemium—Web, API
Mistral AIFreemium$14.99/moWeb, API, CLI, Self-hosted
OnitsAIContact sales—Self-hosted
OpenAI APIPaid$0.1, usage-basedAPI, Web
OpenLITOpen source—Web, CLI, API, Self-hosted
  1. Cohere

    Enterprise AI company offering Command, Embed and Rerank models by API, the North agent platform and private deployments.

  2. Gemini API

    Google's developer API for calling Gemini models from your own apps, with a free tier, pay-as-you-go pricing and Google AI Studio.

  3. Mistral AI

    AI company offering open-weight models, the Vibe assistant, Studio and an API, with a free plan and paid tiers.

  4. OnitsAI

    On-premises AI platforms for managing GPU and NPU capacity (Orbit) and running private LLMs and agents inside your network (Probe).

  5. OpenAI API

    Developer API for OpenAI's GPT models, image generation, realtime voice and transcription, priced per million tokens with batch and flex options.

  6. OpenLIT

    Open-source, OpenTelemetry-native platform for tracing, evaluating and guarding AI agents, with cost and GPU monitoring

Claude Platform FAQ

How much does Claude Platform cost?

Pricing is per million tokens and depends on the model. Claude Haiku 5.5 is $0.10 / MTok for input and $0.50 / MTok for output on prompts up to 100,000 tokens, Sonnet 5.5 is $2 and $10, Opus 5.5 is $4 and $20, and Fable 5.1 is $10 and $50.

Is there a free plan for the Claude API?

Anthropic's pricing pages list no free tier for the developer API. The Claude chat apps have a Free plan at $0, but that is a separate product from the developer platform.

What does batch processing save?

According to Anthropic, Batch API requests are priced at half the rate of standard API calls. They suit large volumes of requests that can run asynchronously.

Can I use Claude through Amazon Bedrock or Google Cloud?

Yes. The docs list the Claude API, Amazon Bedrock, Claude Platform on AWS, Claude Models on Google Cloud and Microsoft Foundry as platforms, and feature support varies by platform.

Does Claude Platform have an API for managing usage?

Yes. The documentation points to an Admin API, a Usage and Cost API and a Compliance API for administration and governance.

What are alternatives to Claude Platform?

Alternatives include the OpenAI API, Google's Gemini API, Mistral AI and Cohere.

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

  • Logo: from claude.com · fetched · Vendor trademark, used only to identify the product; removed on request
  • Cover: from platform.claude.com · fetched · Vendor marketing image from the official site; removed on request

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.