| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-aime-2025 | PRM-CoT (Process-Aware) (Qwen2.5-Math-7B) | – | Accuracy: 20.0 |
| mathematical-reasoning-on-aime24 | PRM-CoT (Process-Aware) (Qwen2.5-Math-7B) | – | Acc: 30.0 |
| mathematical-reasoning-on-math-500 | PRM-CoT (Process-Aware) (Qwen2.5-Math-7B) | – | Accuracy: 85.4 |
| mathematical-reasoning-on-minerva-math | PRM-CoT (Process-Aware) (Qwen2.5-Math-7B) | – | Accuracy: 40.1 |
| mathematical-reasoning-on-olympiadbench | PRM-CoT (Process-Aware) (Qwen2.5-Math-7B) | – | Accuracy: 42.7 |