Coding Agent on SWE-bench Verified (single-run, single-patch, test)
51.6Resolution Rate
Qwen3-Coder
Updated 15d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.05 | 51.6 | - | - | - | |
2026.05 | 41.8 | 1 | 1.6 | 0 | |
2026.05 | 40.2 | 0.92 | - | - | |
2026.05 | 39.4 | - | - | - | |
2026.05 | 39.4 | 1.46 | 2.6 | 0 | |
2026.05 | 39.1 | 2.18 | - | 0 | |
2026.05 | 36.4 | - | - | - | |
2026.05 | 35.1 | 1.15 | 6.6 | 1 | |
| 34.9 | 1.54 | 7.6 | 2 | ||
2026.05 | 26.8 | 1.61 | 12.1 | 3 | |
2026.05 | 20.6 | - | - | - |