Claude

Claude

Usage Based

Anthropic's AI models, from fast Haiku to frontier Fable, built for safe, capable AI at scale.

LLM
Proprietary

Published 29 May 2026Last updated 3 November 2026

Scores

Popularity5/5

Broadly known

Claude is one of the two or three most-referenced LLMs in developer communities worldwide. Massive adoption via Claude.ai, API, Bedrock, and Vertex AI. Powers Claude Code, Cursor (optional model), and many third-party AI products.

Learning Curve2/5

Easy

The Anthropic API follows the standard messages array pattern familiar from other LLM providers. Python and TypeScript SDKs are well-documented. Basic integration takes under an hour.

Flexibility4/5

Flexible

Supports tool use, vision, extended thinking, prompt caching, batch processing, streaming, and system prompts. No fine-tuning and no image output limits the ceiling. Multi-cloud availability adds operational flexibility.

Performance5/5

Best-in-class

Claude 4 Opus and Sonnet 4 are at the frontier of reasoning and coding benchmarks as of 2025. Extended thinking mode significantly improves accuracy on hard problems. Competitive with GPT-4o and Gemini Ultra across most evaluation suites.

Portability3/5

Some lock-in

Anthropic API uses its own message format and tool-calling schema — not OpenAI-compatible. Migrating to/from Claude requires prompt and schema rewrites. Available on Bedrock and Vertex AI reduces infrastructure lock-in but does not eliminate model lock-in.

About Claude

Claude is Anthropic's family of large language models, offered in tiers that trade capability against speed and cost. Claude Fable 5.1 is the most capable widely released model, aimed at demanding reasoning and long-horizon agentic work; Claude Mythos 5.1 is the same model with safeguards tuned for vetted cybersecurity and life-sciences access programs. Claude Opus 5.5 comes close to Fable-level results on most coding and agentic benchmarks at a lower price and higher speed, Claude Sonnet 5.5 is the balanced mid-tier, and Claude Haiku 5.5 is the fast, low-cost entry point. Earlier models such as Fable 5, Opus 5, Sonnet 5, and Haiku 4.5 stay available for existing integrations.

All current models have a 1M-token context window, accept text, images, and PDFs, and support tool use, including computer-use and browser-use tools for automating graphical interfaces. They use adaptive thinking, where the model decides how much to reason and an effort setting steers the depth. Claude Managed Agents hosts long-running agents on Anthropic's infrastructure with session-based billing, and Claude Code, Anthropic's coding agent, runs on the same models.

Access is through the Claude API, Amazon Bedrock, Google Cloud, and Microsoft Foundry, as well as the claude.ai apps. Pricing is per token and varies by tier, with Haiku priced by prompt length, prompt caching that bills cache hits at a small fraction of the input price, a 50% Batch API discount for asynchronous work, and a fast mode on Opus models for lower latency at a higher rate.

Key Features

  • Claude Fable 5.1: Anthropic's most capable widely released model
  • Claude Opus 5.5: near-Fable performance at lower cost and higher speed
  • Claude Sonnet 5.5: balanced mid-tier for most production work
  • Claude Haiku 5.5: low-cost entry tier with a 1M-token context window
  • Adaptive thinking with adjustable effort levels across the current lineup
  • Computer-use and browser-use tools for GUI automation
  • Prompt caching, a 50% Batch API discount, and Managed Agents hosting
  • Available on the Claude API, Bedrock, Google Cloud, and Microsoft Foundry

Pros

  • 1M-token context across the current lineup makes whole-repository and whole-document reasoning practical
  • Adaptive thinking with an effort setting trades speed for accuracy on hard multi-step problems
  • Computer use enables browser and GUI automation without custom RPA tooling
  • Prompt caching sharply cuts cost for applications with long, repeated system prompts
  • Available on Bedrock, Google Cloud, and Foundry, so it fits existing cloud accounts
  • Tiered lineup lets you match cost to task without switching providers

Cons

  • No image generation or audio: text, image, and PDF input only
  • No fine-tuning through the Anthropic API
  • Top-tier output tokens cost more than some competitors
  • Computer use is still maturing, and reliability on complex GUI workflows varies
  • Rate limits on lower API tiers can throttle production workloads
  • Tool-calling schemas and thinking APIs are Anthropic-specific, which adds switching cost

Claude Pricing

Usage Based
Claude Managed Agents$0.08/usage_based
  • $0.08 per session-hour plus standard token rates
  • Hosted, long-running agent sessions
Claude Mythos 5.1 (limited availability)Contact sales
  • $10/M input, $50/M output, same as Fable 5.1
  • Fable 5.1 with safeguards for vetted cybersecurity and life-sciences programs
  • 1M token context
Claude Fable 5Contact sales
  • $10/M input, $50/M output
  • Previous-generation flagship, still available
  • 1M token context
Claude Opus 5.5Contact sales
  • $4/M input, $20/M output
  • Near-Fable 5.1 performance, faster and cheaper than Opus 5
  • Fast mode: $8/M input, $40/M output; 1M context
Claude Opus 5Contact sales
  • $5/M input, $25/M output
  • Previous-generation Opus tier, still available
  • Fast mode (research preview): $10/M input, $50/M output; 1M context
Prompt CachingContact sales
  • Cache hits: 10% of base input price on most models
  • 5% on Opus 5.5 and Sonnet 5.5; 2.5% on Fable 5.1 and Mythos 5.1
  • 5-minute cache write at 1.25x base price, or 1-hour write at 2x
Batch APIContact sales
  • 50% discount on input and output tokens across every tier
  • Asynchronous processing for non-urgent workloads
Claude Fable 5.1Contact sales
  • $10/M input, $50/M output
  • Most capable widely released model
  • 1M context; cache hits at 2.5% of base input price
Claude Sonnet 5.5Contact sales
  • $2/M input, $10/M output; cache hits $0.10/M
  • Balanced mid-tier: the best mix of speed and intelligence
  • 1M context, 128K max output
Claude Haiku 5.5Contact sales
  • $0.10/M input, $0.50/M output for prompts up to 100K tokens
  • $0.50/M input, $2.50/M output for prompts over 100K tokens
  • Low-cost entry tier; 1M context, 128K max output
Claude Sonnet 5Contact sales
  • $2/M input, $10/M output
  • Previous-generation Sonnet tier, still available
  • 1M token context

Tech Stacks with Claude

OpenClaw Self-Hosted

Infrastructure

Run OpenClaw, the open-source personal AI agent, in Docker on your own server or home hardware. It answers in the messaging apps you already use, thinks with a model such as Claude or OpenAI, and stays closed to the internet, with its dashboard reached over an SSH tunnel.

LLM:
Deploy on:
Model Inference add-on:
Model Aggregator add-on:
Tunnel add-on:
Reverse Proxy add-on:
Self-Hosted PaaS add-on:

n8n AI Agent

Project

Build an AI agent in n8n's visual editor. The AI Agent node connects a model such as Claude or OpenAI to tools, memory, and a chat window, and the agent decides which tool to call for each request. The tools are n8n's own integrations, your other workflows, and MCP servers, so there is no backend code to write.

LLM:
Database add-on:
Vector Database add-on:
Model Inference add-on:
Model Aggregator add-on:

Devin Dev Stack

Developer

Devin Desktop as the editor and agent manager, with Devin Cloud for work that keeps going without your machine, Claude as the default model and OpenAI, Gemini, or Grok in the same picker, plus your pick of GitHub or GitLab for version control.

LLM:
Version Control:
Coding Agent add-on:
Code Review add-on:
CI/CD add-on:
Containerization add-on:

Tools Related to Claude

Integrates with Claude(16)

Claude Code is Anthropic's own CLI, natively built for Claude models, though other providers can be wired in via configuration.

Cursor offers Claude models, including the current Sonnet, Opus, and Fable releases, in its model picker for Agent and chat. Picking one manually draws on the usage included with the Cursor plan.

Claude models are one of the model families available in Devin on paid plans; Devin supplies the IDE, Devin Local agent, and Agent Command Center around them.

Claude is used as the AI reasoning engine in n8n workflows via the Anthropic node.

Claude is used in Zapier workflows via the Anthropic action.

Claude is available in Make automation scenarios via the Anthropic module.

Alternatives to Claude(5)

Claude and OpenAI are the two leading proprietary LLM APIs, each with a tiered lineup and 1M-token context on its main models. OpenAI's platform also covers image generation, realtime voice, transcription, and embeddings; Claude outputs text only and is the model family behind Claude Code and Anthropic's computer-use tools.

Google Gemini takes audio and video input alongside text and images, grounds answers in Google Search, and has a free tier in AI Studio. Claude accepts text, images, and PDFs only and is usually picked for agentic coding and computer use. Both offer 1M-token context on their top models.

Meta Llama is an open-weight family you self-host or call through third-party providers, with no per-token cost on your own hardware. Claude is a managed proprietary API that leads on the hardest reasoning and agentic coding tasks but cannot be self-hosted.

Muse Spark offers a 1M-token context window, video input, and a discounted Contributor tier through Meta's OpenAI-compatible API; Claude has the longer track record with developers and far wider availability, since Meta's own API is still a US-only preview.

Grok searches X posts in real time and adds Grok Imagine image and video generation, with an OpenAI-compatible API. Claude generates text only and is usually chosen for enterprise deployments, agentic coding, and its availability on Bedrock, Vertex AI, and Foundry.

Vendor

Anthropic

Anthropic

Website

Tags

Web

Details

Maintained
Yes