| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-aime-2025 | Qwen2.5-7B-Instruct (OT3, dense) | – | Accuracy: 54.0 |
| mathematical-reasoning-on-aime24 | Qwen2.5-7B-Instruct (OT3, dense) | – | Acc: 61.0 |
| mathematical-reasoning-on-math-500 | Qwen2.5-7B-Instruct (OT3, dense) | – | Accuracy: 95.0 |