Issue Resolution on poetry GitHub Issues OSS repo (10 randomly selected)
4Issues ResolvedAIR
Evaluation results
| Method | Links | |
|---|---|---|
| AIRwall-clock budget=same, tool budget=same2025.12 | 4 | |
| Gemini-2.5 prowall-clock budget=same, tool budget=same2025.12 | 3 | |
| Claude 4 Sonnetwall-clock budget=same, tool budget=same2025.12 | 1 |