| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-minerva-math | GRAIL (Qwen3-8B) | #18 | Accuracy: 44.73 |
| mathematical-reasoning-on-olympiadbench | GRAIL (Qwen3-8B) | #27 | Accuracy: 49.17 |
| mathematical-reasoning-on-math-500 | GRAIL (Qwen3-8B) | #49 | Accuracy: 92.07 |
| mathematical-reasoning-on-aime24 | GRAIL (Qwen3-8B) | #84 | Acc: 47.78 |