OpenEvals - SWE-bench Pro: leaderboard
Hugging Face OpenEvals official leaderboard covering SWE-bench Pro software issue resolution.
Metric: Resolved (%). Source: huggingface.co. 11 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3-Coder-Next | 44.3 | |
| 2 | Qwen3-Coder-480B-A35B-Instruct | 38.7 | |
| 3 | MiniMax-M2.1 | 36.81 | #289 |
| 4 | Kimi-K2-Instruct | 27.67 | |
| 5 | Qwen3-235B-A22B | 21.41 | |
| 6 | gpt-oss-120b | 16.2 | |
| 7 | DeepSeek-V3.2 | 15.56 | |
| 8 | gemma-3-27b-it | 11.38 | |
| 9 | Llama-3.1-405B-Instruct | 11.18 | |
| 10 | GLM-4.6 | 9.67 | #252 |
Interactive version: theaggregate.ai/benchmark?slug=openevals-swe-bench-pro · How It Works · Data refreshed daily, snapshot 2026-10-20.