| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-aime-2025 | HDL (Qwen3-8B) | – | Accuracy: 70.56 |
| mathematical-reasoning-on-aime-2026 | HDL (Qwen3-8B) | – | Accuracy: 69.1 |
| mathematical-reasoning-on-aime24 | HDL (Qwen3-8B) | – | Acc: 75.97 |
| mathematical-reasoning-on-hmmt-feb-2026 | HDL (Llama3.1-8B) | – | Accuracy: 29.04 |
| mathematical-reasoning-on-minerva-math | HDL (Qwen3-8B) | – | Accuracy: 32.49 |
| mathematical-reasoning-on-olympiadbench | HDL (Llama3.1-8B) | – | Accuracy: 62.95 |