Trajectory Verification on SWE-bench Verified Qwen3-Coder-Flash trajectories
62RM@32
SWE-RM-30A3B
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2025.12 | 62 | 0.783 | 0.051 | |
2025.12 | 54.6 | - | - | |
2025.12 | 53.2 | 0.758 | 0.124 | |
2025.12 | 52.6 | - | - | |
2025.12 | 51.2 | 0.776 | 0.223 |