Claude Sonnet 3.7: benchmarks and ranking
Claude Sonnet 3.7 (Anthropic) — overall rank: #117 / 153, overall score: 24.8 (90% interval 19.1–31.6), leaderboards: 4. Released: 2025-02-24 · Price (in/out): $3 / $15
Rank on each leaderboard
| Leaderboard | Rank | Published value |
|---|---|---|
| Artificial Analysis | Not rated | — |
| LiveBench | Not rated | — |
| Epoch AI | #113 / 230 | 141.03 |
| ARC Prize (ARC-AGI) | #81 / 95 | 0.9 |
| Scale AI SEAL | #32 / 44 | 8.04 |
| Vals AI | Not rated | — |
| Terminal-Bench | Not rated | — |
| EQ-Bench | Not rated | — |
| SimpleBench | #70 / 112 | 46.4 |
| BenchLM.ai | Not rated | — |
Domain scores
- Coding: 41.3
- Reasoning: 32.0
- Math: 38.7
- Knowledge: 51.6
- Writing: 43.5
- Instructions: 30.0
- Chat & EQ: 22.8
- Business: 56.0
- Japanese: 55.7
- Vision: 23.0