| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-aime-2025 | Cal-OPD (Qwen3-30B-A3B-Thinking-2507 -> Qwen3-4B) | – | Accuracy: 61.3 |
| mathematical-reasoning-on-aime-2026 | Cal-OPD (Qwen3-30B-A3B-Thinking-2507 -> Qwen3-4B) | – | Accuracy: 60.2 |
| mathematical-reasoning-on-aime24 | Cal-OPD (Qwen3-30B-A3B-Thinking-2507 -> Qwen3-4B) | – | Acc: 67.1 |
| mathematical-reasoning-on-math-500 | Cal-OPD (Qwen3-30B-A3B-Thinking-2507 -> Qwen3-4B) | – | Accuracy: 95.8 |