Multimodal
SWE-bench · 22 models
| Rank | Model | Score |
|---|---|---|
| 1 | Codefuse_Pycfuse_SVR | 35.98 |
| 1 | GUIRepair + o3 (2025-04-16) | 35.98 |
| 3 | Refact.ai Agent | 35.59 |
| 4 | OpenHands-Versa (Claude-Sonnet 4) | 34.43 |
| 5 | GUIRepair + o4-mini (2025-04-16) | 33.85 |
| 6 | OpenHands-Versa (Claude-3.7 Sonnet) | 31.33 |
| 7 | GUIRepair + GPT 4.1 (2025-04-14) | 31.14 |
| 8 | Zencoder (2025-04-01) | 30.56 |
| 9 | GUIRepair + GPT 4o (2024-08-06) | 30.37 |
| 10 | Globant Code Fixer Agent | 29.59 |
| 11 | Zencoder (2025-03-10) | 27.08 |
| 12 | Agentless Lite + Claude-3.5 Sonnet | 25.34 |
| 13 | SWE-agent + Claude Sonnet 3.5 | 12.19 |
| 13 | SWE-agent Multimodal + GPT 4o (2024-08-06) | 12.19 |
| 15 | SWE-agent + GPT 4o (2024-08-06) | 11.99 |
| 15 | SWE-agent JavaScript + Claude Sonnet 3.5 | 11.99 |
| 17 | SWE-agent Multimodal + Claude 3.5 Sonnet | 11.41 |
| 18 | SWE-agent JavaScript + GPT 4o (2024-08-06) | 9.28 |
| 19 | Agentless + Claude 3.5 Sonnet | 6.19 |
| 20 | RAG + GPT 4o (2024-08-06) | 6 |
| 21 | RAG + Claude 3.5 Sonnet | 5.03 |
| 22 | Agentless + GPT 4o (2024-08-06) | 3.09 |