Task Performance on SWE-bench (Success, Turns)
71.9Success RateClaude Opus 4.7
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Claude Opus 4.72026.05 | 71.9 | 12.12 | |
| Claude Sonnet 4.62026.05 | 68.8 | 20.66 | |
| Gemini 3.1 Pro Preview2026.05 | 68.8 | 37.36 | |
| GPT-5.2 Instant2026.05 | 57.8 | 21.52 | |
| Qwen3 235B2026.05 | 33.3 | 62.92 |