Claude 3.7 Sonnet (64k thinking): benchmark scores, price and ranking
Claude 3.7 Sonnet (64k thinking) from Anthropic ranks #193 of 743 LLMs on the LLMs Tiger aggregate benchmark score (50.4), based on 2 benchmark results captured up to 2026-10-13.
- Organisation
- Anthropic
- Type
- Frontier model
- Open weights
- No
- Released
- 2025-02-24 (source: curated)
- Cheapest API price
- $3.00 in / $15.00 out per million tokens (via replicate, observed 2026-10-13T20:47Z)
- Aggregate rank
- #193 of 743 (score 50.4)
Across the 2 benchmarks where at least ten models are scored, Claude 3.7 Sonnet (64k thinking) is in the top 10% on 0 and in the top half on 1.
Its cheapest listed input price, $3.00 per million tokens, is lower than that of 11% of the 427 models with a listed price.
Claude 3.7 Sonnet (64k thinking) benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| GPQA Diamond | 79.7% | #106 of 275 | Epoch AI Benchmarking Hub | 2026-10-13 |
| OTIS Mock AIME | 57.8% | #95 of 189 | Epoch AI Benchmarking Hub | 2026-10-13 |
About Claude 3.7 Sonnet (64k thinking)
Claude 3.7 Sonnet (64k thinking) is a frontier model from Anthropic. It was released on 2025-02-24, according to the developer's announcement. The cheapest API listing we track charges $3.00 per million input tokens and $15.00 per million output tokens, through replicate. We hold 2 benchmark results for it across 1 area, and none is in the top tenth of its board.
Where Claude 3.7 Sonnet (64k thinking) stands
On our aggregate score, Claude 3.7 Sonnet (64k thinking) is #193 of 743 ranked models, around the middle of the table. Its score is 50.4; the scale is centred on 50, so it sits 0.4 points above the typical model. That beats 74% of the ranked field. Among the 70 ranked Anthropic models it is #32. The next one up is swe-search at 50.5. The next one down is Claude 4.5 Haiku at 50.3.
Claude 3.7 Sonnet (64k thinking), benchmark by benchmark
GPQA Diamond puts Claude 3.7 Sonnet (64k thinking) at 79.7%, #106 of 275. That beats 61% of the models tested, but 16.0 points short of GPT-6 Astra (max) (95.8%). The results either side are Qwen3-235B-A22B-Thinking-2507 (80.1%) and o4-mini (high) (79.6%). It covers expert-level science questions. The board median is 71.7%, so Claude 3.7 Sonnet (64k thinking) is 8.0 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Claude 3.7 Sonnet (64k thinking) lands below the middle on OTIS Mock AIME: 57.8%, #95 of 189, with 50% of the field scoring higher. Claude Fable 5.1 (max) leads at 100.0%; the closest result above is glm-4.7-flash at 58.3%. It covers competition maths (AIME 2024-25 style). The median score on this board is also 57.8%. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
Reasoning, maths and science: solid, ahead of 56% of the field on average across 2 benchmarks. Its best board here is GPQA Diamond (#106 of 275) and its weakest is OTIS Mock AIME (#95 of 189).
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Claude 3.7 Sonnet (64k thinking) costs more than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1), GLM-5.3-Flash (max) ($0.026, 49.6). Similar scores cost more on Muse Wonder (max) ($4.00) and gpt-4o-2024-05-13 ($5.00).
Release history
The Anthropic model released before it was claude-3-7-sonnet-20250219 on 2025-02-24, which scores 48.2. The next one was claude-sonnet-4-20250514_12K on 2025-05-14, at 49.9.
Reading these numbers
The 2 scores on this page come from 1 source: Epoch AI Benchmarking Hub (2). No official leaderboard supplied any of them, so treat them as third-party measurements. The leaderboard also lists 7 other reasoning-effort settings of this model, which redirect here. This page shows the highest. With 2 results the aggregate rank can move a lot when one new score arrives.
Claude 3.7 Sonnet (64k thinking) is also listed at 7 other reasoning-effort settings; this page shows the highest-effort row, which is the one the leaderboard keeps by default.
Models ranked nearby
- #190 gemini-2.5-pro_16K (50.4)
- #191 nousresearch/hermes-3-llama-3.1-405b (50.4)
- #192 ollama/Qwen2.5.1-Coder-7B-Instruct-GGUF:Q8_0-32k (50.4)
- #194 Nemotron-Cascade-2-30B-A3B (50.4)
- #195 DeepSeek-V3.2-Speciale (50.4)
- #196 gemini-2.5-flash-preview-04-17 (50.4)
More from Anthropic
- Claude Opus 5.5 (max) (#1, 64.3)
- Claude Fable 5.1 (max) (#3, 61.9)
- Claude Sonnet 5.5 (max) (#4, 60.8)
- Claude Opus 5 (max) (#8, 60.0)
- Claude Fable 5 (max) (#9, 59.9)
- claude-opus-4-8-xhigh-effort (#14, 56.6)
Compare it on the live leaderboard → · How the aggregate score works