| Benchmark | Model | Rank | Results |
|---|---|---|---|
| arithmetic-reasoning-on-gsm8k | Qwen3-8B-Base (R-Few 1%) | – | Accuracy: 94.0 |
| mathematical-reasoning-on-math-500 | Qwen3-8B-Base (R-Few 5%) | – | Accuracy: 82.6 |
| mathematical-reasoning-on-minerva-math | Qwen3-8B-Base (R-Few 5%) | – | Accuracy: 60.3 |
| mathematical-reasoning-on-olympiadbench | Qwen3-8B-Base (R-Few 5%) | – | Accuracy: 46.4 |
| multi-task-language-understanding-on-mmlu-pro | Qwen3-8B-Base (R-Few 5%) | – | Accuracy: 63.2 |