Anthropic
Reasoning

Claude Sonnet 3.7

Previous Gen

The first Claude with hybrid reasoning, a single model that can answer instantly or "think harder" via extended thinking.

Claude Sonnet 3.7 is a reasoning AI model from Anthropic. It costs $3.00 per million input tokens and $15.00 per million output tokens (blended $6.60/M), with a 200,000-token context window.

RatesUSD per million tokens. Blend is 70% input, 30% output, our calculation.

INPUT
$3.00/M
per million input tokens
OUTPUT
$15.00/M
per million output tokens
BLENDED 70/30
$6.60/M
unchanged since 3 May
CONTEXT
200,000
tokens

Hand-maintained from Anthropic's published pricing; no check date on record.

What it is good at
  • Hybrid extended thinking
  • Coding strength
  • Faithful instruction following
Typical use cases
  • Reasoning-heavy chat
  • Code review

Benchmarks

vs. best public score
Graduate-level science questions, "Google-proof". Source: Anthropic (self-reported), GPQA Diamond, read 2026-11-10.
Real GitHub issues solved end-to-end. Source: Anthropic (self-reported), SWE-bench Verified, read 2026-11-10.
LMArena Elo1372 Elo
Crowd-sourced head-to-head preference Elo rating. Source: LMArena (independent), LMArena Text Arena score, read 2026-11-10.
Hand-curated from each provider's published reports and public leaderboards. Methodology varies across sources, treat as directional rather than authoritative.

Also reported by the provider

Results on benchmarks outside the set we compare across models. They are listed for completeness and are deliberately excluded from the bars above and from every ranking on this site, because a score is only comparable against the same benchmark variant. Why benchmark scores do not compare →

  • MATH-50096.2%self-reported by Anthropic · 2025-02-24 · 64K extended thinking (82.2% without). Pass@1 averaged over several trials. Not comparable with the MATH column, so shown here and never ranked.
  • MMMLU83.2%self-reported by Anthropic · 2025-02-24 · Launch table, no extended thinking (86.1% with 64K extended thinking). Pass@1 averaged over several trials. Not comparable with the MMLU column, so shown here and never ranked.

More from Anthropic

See all 18 →

Frequently asked questions

How much does Claude Sonnet 3.7 cost?

Claude Sonnet 3.7 costs $3.00 per million input tokens and $15.00 per million output tokens, for a blended reference rate of $6.60 per million tokens.

What is Claude Sonnet 3.7's context window?

Claude Sonnet 3.7 supports up to 200,000 tokens of context in a single request.

What is Claude Sonnet 3.7 best for?

Claude Sonnet 3.7 is well suited to Hybrid extended thinking, Coding strength and Faithful instruction following.

Who makes Claude Sonnet 3.7?

Claude Sonnet 3.7 is developed and served by Anthropic. It was released in Feb 2025.

Terms used on this page