Anthropic

Anthropic is an AI safety company and creator of the Claude family of large language models (Opus, Sonnet, Haiku). The Claude API provides access to Claude models for text generation, vision, tool use, extended thinking, batch processing, and agentic workflows including managed agents, skills, memory, compaction, and computer use. Anthropic also publishes the open Model Context Protocol (MCP) for standardized AI tool integration and ships Claude Code, the terminal-based agentic coding tool.

11 APIs 26 Features

AIArtificial IntelligenceClaudeFoundation ModelsLarge Language ModelsMachine LearningMCPAgents

APIs

Anthropic Messages API

Send a structured list of input messages with text and/or image content, and the model will generate the next message in the conversation. The Messages API supports text, images...

Anthropic Models API

List and inspect Claude models including Opus 4.7, Sonnet 4.6, and Haiku 4.5. The response includes max_input_tokens, max_tokens, and a capabilities object for every model so cl...

Anthropic Message Batches API

Send a batch of Message creation requests at 50% off both input and output tokens. Batches can take up to 24 hours to complete. Supports up to 300k output tokens per request on ...

Anthropic Files API

The Files API lets you upload and manage files for reuse across Messages, Batches, code execution, and Managed Agents without re-uploading content. 500 MB request limit; support...

Anthropic Admin API

Programmatically manage administrative resources including organization info, members, invites, workspaces, workspace members, and API keys. Requires an Admin API key (sk-ant-ad...

Anthropic Prompts API

Prompt tools API for generating, improving, and templatizing prompts. Backs the Console Prompt Generator and integrates into the Workbench prompt authoring workflow.

Anthropic Token Counting API

Count the number of tokens in a Message, including tools, images, and documents, without creating it. The Token Count API accepts the same structured list of inputs as Messages ...

Anthropic Skills API

Create and manage custom Agent Skills. Skills are filesystem-based directories of instructions, scripts, and resources that Claude loads on demand via progressive disclosure. Wo...

Anthropic Usage and Cost API

Programmatically access your organization's API usage and cost data with the Usage & Cost Admin API. The Usage API tracks token consumption with breakdowns by model, workspace, ...

Anthropic Claude Code Analytics API

Daily aggregated user-level Claude Code productivity metrics — sessions, lines of code, commits, pull requests, tool actions, and estimated costs by model — via /v1/organization...

Anthropic Managed Agents API

Claude Managed Agents — a fully managed agent harness for running Claude as an autonomous agent with secure sandboxing, built-in tools (bash, file ops, web search/fetch, MCP), S...

ARAZZO

Anthropic Upload, Verify, and Download a File

Upload a file, read back its metadata, and download its content when it is downloadable.

ARAZZO

GraphQL

Anthropic GraphQL Schema

This GraphQL schema is a conceptual representation of the Anthropic Claude API derived from the REST API surface. The Anthropic API provides access to Claude large language mode...

GRAPHQL

Pricing Plans

Anthropic Plans Pricing

5 plans

PLANS

Rate Limits

Anthropic Rate Limits

12 limits

RATE LIMITS

FinOps

Anthropic Finops

FINOPS

Features

Claude Opus 4.7 — most capable generally available model for complex reasoning and agentic coding

Claude Sonnet 4.6 — balanced model combining speed and intelligence with 1M context window

Claude Haiku 4.5 — fastest model with near-frontier intelligence

Messages API with text, vision, tool use, extended thinking, streaming, structured outputs

Prompt caching with automatic and manual breakpoints, 5-minute and 1-hour TTLs, cache reads at 10% of input price

Server-side compaction API for effectively infinite conversations on Opus 4.6/4.7 and Sonnet 4.6

Memory tool (client-side) for cross-conversation persistence with ZDR support

Message Batches API with 50% discount on both input and output tokens

Files API for upload-once / reuse-many content across Messages and Managed Agents

Agent Skills API (beta) — packaged domain expertise with progressive disclosure

Claude Managed Agents (beta) — Agents, Sessions, Environments with SSE streaming and managed sandboxes

Web Search tool ($10/1,000 searches) and Web Fetch tool (free)

Code Execution tool — Python + Bash + filesystem in sandbox; 1,550 free container-hours/month

Computer Use tool for browser and desktop automation (beta)

Advisor tool (beta) for pairing executor and high-intelligence advisor models

Tool Search tool and Programmatic Tool Calling

Token-bucket rate limiting with cache-aware ITPM (cache reads excluded on most models)

Five usage tiers (Tier 1-4 plus Monthly Invoicing) with automatic advancement

Usage & Cost Admin API and Claude Code Analytics Admin API for FinOps reporting

Rate Limits API for programmatic limit inspection

Workload Identity Federation for short-lived bearer tokens

Data residency controls (US-only inference at 1.1x pricing for models after Feb 2026)

Available via Claude API, Claude Platform on AWS, Microsoft Foundry, AWS Bedrock, and Google Vertex AI

{"Official SDKs" => "Python, TypeScript, Java, Go, Ruby, C#, PHP plus the `ant` CLI"}

Claude Code (terminal agent) and Claude Code GitHub Action

Open MCP specification stewardship via modelcontextprotocol.io

Anthropic

Sources