CrossModel
Claude Sonnet 4.6
anthropic/claude-sonnet-4-6
Back to model catalog

Anthropic · Model guide

Claude Sonnet 4.6

Compare
anthropic/claude-sonnet-4-6

The February 2026 Sonnet: a full upgrade across coding, computer use, long-context reasoning and design, and the first Sonnet with a 1M token context window. It still accepts the deprecated extended-thinking parameters alongside adaptive thinking, and its effort ladder stops short of xhigh. Sonnet 5 has since replaced it.

Modalities
TextImageText
Context
1M
Max output
128K
Price / 1M tokens
$ 3.00in$ 15.00out
StreamingTool callingJSON outputImage inputThinking mode · can be turned off

What this model is good at

What the vendor positions it for, and which of our models to reach for instead.

VendorLegacy model: still available, with migration to a current model recommended.source

Best for

  • Agentic coding · Anthropic recommends medium effort as the default for agentic coding and tool-heavy workflows on this model.(vendor claim)source

  • Long-context work · The launch introduced a 1M token context window for the Sonnet tier.(vendor claim)source

Neighbouring models
Compare them side by side →

Pricing and billing

Billed per token. Cached input is charged at the cache-read rate.

Price / 1M tokensYou pay
Input$ 3.00
Output$ 15.00
Cache read$ 0.30
Cache write$ 3.75
Estimate a request
$ 0.0600
Estimated cost per request
10.0K × $ 3.00
+ 2.0K × $ 15.00
≈ $ 60.00 per 1,000 requests

Capabilities and limits

What CrossModel guarantees across every route this model can take right now.

Context window
1.0M tokens
Max output
128.0K tokens
Input / output modalities
Text + Image → Text
Streaming
Supported
Tool calling
Supported
Structured output (JSON)
Supported
Image input
Supported
Thinking mode
Supported· can be turned off
Reasoning effort
low · medium · high · max
Thinking budget
1.0K – 0 tokens
Available endpoints
/v1/chat/completions · /v1/responses · /v1/messages
First-party sources:Specsplatform.claude.comCapabilitiesplatform.claude.com 1platform.claude.com 2platform.claude.com 3platform.claude.com 4

Published benchmarks

Scores the vendor reported, with the evaluation setup each one came from.

Agentic coding
  • Terminal-Bench 2.0
    pass_rate·Anthropic · 2026-02-17setup not fully disclosed
    59.1%
Agentic search
  • BrowseComp (with tools)
    accuracy·Anthropic · 2026-02-17setup not fully disclosed
    74.7%
Coding
  • SWE-Bench Verified
    resolved_rate·Anthropic · 2026-02-17setup not fully disclosed
    79.6%
Computer use
  • OSWorld-Verified
    success_rate·Anthropic · 2026-06-30setup not fully disclosed

    Anthropic changed how it runs OSWorld-Verified and restated this score in the Sonnet 5 launch post; the Sonnet 4.6 launch post reported 72.5%.

    78.5%
Knowledge work
  • GDPval-AA
    elo·Anthropic · 2026-02-17setup not fully disclosed
    1633
Multilingual
  • MMMLU
    accuracy·Anthropic · 2026-02-17setup not fully disclosed
    89.3%
Multimodal reasoning
  • MMMU Pro (no tools)
    accuracy·Anthropic · 2026-02-17setup not fully disclosed
    74.5%
Reasoning
  • ARC-AGI-2
    score·Anthropic · 2026-02-17setup not fully disclosed
    58.3%
  • Humanity's Last Exam
    accuracy·Anthropic · 2026-06-30setup not fully disclosed

    No tools. Anthropic changed the grader model and restated this score in the Sonnet 5 launch post; the Sonnet 4.6 launch post reported 33.2%.

    34.6%
  • Humanity's Last Exam (with tools)
    accuracy·Anthropic · 2026-06-30setup not fully disclosed

    Anthropic changed the grader model and restated this score in the Sonnet 5 launch post; the Sonnet 4.6 launch post reported 49.0%.

    46.8%
Scientific reasoning
  • GPQA Diamond
    accuracy·Anthropic · 2026-02-17setup not fully disclosed
    89.9%

Vendor-reported numbers, not CrossModel measurements. Scores are only comparable when the benchmark version, metric and evaluation setup match, so nothing here is averaged or ranked.

Use it in your tools

Point the base URL at CrossModel and paste this model ID — every tool below has a setup guide.

Claude Code
Anthropic's official command-line coding assistant
Cursor
AI-first code editor
Cline
Open-source AI coding agent for VS Code
Codex
OpenAI's coding agent for ChatGPT desktop and the terminal
OpenCode
Open-source terminal coding assistant
Zed
High-performance, AI-native code editor
Open WebUI
Open-source, self-hosted AI chat interface
Continue
Open-source AI coding assistant for VS Code & JetBrains
See all integration guides →

Frequently asked questions

What is Claude Sonnet 4.6?
Claude Sonnet 4.6 is available on CrossModel as anthropic/claude-sonnet-4-6. It has a 1M-token context window and can return up to 128K tokens per request.
How much does Claude Sonnet 4.6 cost?
Currently $ 3.00 per 1M input tokens and $ 15.00 per 1M output tokens. Cached input is billed at $ 0.30 per 1M.
Does Claude Sonnet 4.6 support tool calling and structured output?
Tool calling: Supported · Structured output (JSON): Supported · Image input: Supported · Streaming: Supported · Thinking mode: Supported
Which endpoint do I call?
OpenAI Chat Completions (/v1/chat/completions), OpenAI Responses (/v1/responses) and Anthropic Messages (/v1/messages) all work — pick whichever your SDK already speaks.
Can I try it without writing code?
Yes. Sign in and open the Playground on this page to chat with the model directly — no API key needed. Usage is billed from your wallet at the normal rate.

Chat first, integrate later

Chat first, then wire it in once you like the answers.