| Benchmark | Model | Rank | Results |
|---|---|---|---|
| code-generation-on-swe-bench-verified | OpenSWE-72B (SWE-Agent) | #54 | Accuracy: 66.0 |
| code-generation-on-swe-bench-verified | OpenSWE-32B (SWE-Agent) | #58 | Accuracy: 62.4 |
| multi-task-language-understanding-on-mmlu-pro | OpenSWE-72B (Qwen2.5-72B-Base) | #64 | Accuracy: 72.7 |
| code-generation-on-humaneval | OpenSWE-72B (Qwen2.5-72B-Base) | #72 | Pass@1: 76.25 |
| arithmetic-reasoning-on-gsm8k | OpenSWE-72B (Qwen2.5-72B-Base) | #79 | Accuracy: 89.16 |
| mathematical-reasoning-on-math-500 | OpenSWE-72B (Qwen2.5-72B-Base) | #110 | Accuracy: 72.6 |