Model card
Claude 3.7 Sonnet.
Anthropicapi
§ 01 · Benchmarks
Every benchmark Claude 3.7 Sonnet has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | RE-Bench | Agentic AI · RE-Bench | normalized-score | 0.3% | #2 | 2025-04-01 | source ↗ |
| 02 | HCAST | Agentic AI · HCAST | success-rate | 38.0% | #3 | 2025-04-01 | source ↗ |
| 03 | METR Time Horizon | Agentic AI · Time Horizon | task-horizon-minutes | 14.0% | #3 | 2025-04-01 | source ↗ |
| 04 | WebArena | Agentic AI · Web & Desktop Agents | success-rate | 35.1% | #5 | 2025-02-24 | source ↗ |
| 05 | AIME 2024 | Reasoning · Mathematical Reasoning | accuracy | 80.0% | #5 | — | source ↗ |
| 06 | ARC-AGI-1 | Reasoning · Logical Reasoning | accuracy | 30.0% | #5 | — | source ↗ |
| 07 | OSWorld | Agentic AI · Web & Desktop Agents | success-rate | 28.0% | #8 | — | source ↗ |
| 08 | MATH | Reasoning · Mathematical Reasoning | accuracy | 96.2% | #9 | — | source ↗ |
| 09 | SWE-Bench Verified | Computer Code · Code Generation | resolve-rate | 63.7% | #27 | — | source ↗ |
| 10 | SWE-bench Verified | Agentic AI · SWE-bench | resolve-rate | 70.3% | #38 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where Claude 3.7 Sonnet actually performs.
§ 03 · Papers
2 papers with results for Claude 3.7 Sonnet.
- 2025-04-01· Agentic AI· 3 results
METR: Measuring Autonomy in AI Systems (2025 Update)
- 2025-02-24· Agentic AI· 1 result
Claude 3.7 Sonnet System Card
§ 04 · Related models
Other Anthropic models scored on Codesota.
Claude Opus 4
Undisclosed params · 13 results · 2 SOTA
Claude Opus 4.5
3 results · 2 SOTA
Claude Sonnet 5
Undisclosed params · 2 results · 2 SOTA
Claude Sonnet 4
10 results · 1 SOTA
Claude Mythos Preview
1 result · 1 SOTA
Claude 3.5 Sonnet
Undisclosed params · 27 results
Claude Opus 4.5
Undisclosed params · 13 results
Claude 3 Opus
5 results
§ 05 · Sources & freshness
Where these numbers come from.
arxiv
3
results
official-leaderboard
2
results
anthropic-blog
2
results
anthropic-model-card
1
result
arcprize-leaderboard
1
result
editorial
1
result
8 of 10 rows marked verified. · first result 2025-02-24, latest 2025-04-01.