Claude 3.7 Sonnet
- Made by
- Anthropic
- Released
- Announced in
- Claude 3.7 Sonnet and Claude Codeanthropic.com
- Event
- Claude 3.7 Sonnet
- API record
- /api/v1/models/claude-3-7-sonnet
Benchmark results
| Benchmark | Score | Source check |
|---|---|---|
| SWE-bench Verifiedbash and editor tools plus a thinking tool; averaged over 16 trials | 62.3% | CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| internal scoring and a custom scaffold, on a reduced subset of problems · parallel test-time compute | 70.3% | CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.CorroboratedThe same number is printed in 1 other lab document: 1 |
| GPQA Diamondpass@1 averaged over several trials | 78.2% | CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| parallel test-time compute with internal scoring · parallel test-time compute | 84.8% | CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.CorroboratedThe same number is printed in 1 other lab document: 1 |
| no extended thinking | 68.0% | CheckedA person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| AIME 2024No tools · 64K extended thinking; pass@1 averaged over up to 16 trials | 61.3% | CheckedA person checked this number against its source on 18 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| 64K extended thinking; parallel test-time compute with internal scoring · parallel test-time compute | 80.0% | CheckedA person checked this number against its source on 18 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| no extended thinking; pass@1 averaged over up to 16 trials | 23.3% | CheckedA person checked this number against its source on 18 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
Sources
Reported by Anthropic
- Claude 3.7 Sonnet and Claude Codeanthropic.com · 24 Feb 2025
Original image
