Mathematical Reasoning on AIME 2024 (p@1 and p@16)
87.92p@1LED
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LEDModel=Qwen3-30B-A3B-Thinking2026.02 | 87.92 | 90 | |
| DoLaModel=Qwen3-30B-A3B-Thinking2026.02 | 87.5 | 90 | |
| CoTModel=Qwen3-30B-A3B-Thinking2026.02 | 87.29 | 90 | |
| ST-GModel=Qwen3-30B-A3B-Thinking2026.02 | 86.46 | 90 | |
| STModel=Qwen3-30B-A3B-Thinking2026.02 | 85.62 | 90 | |
| STModel=Qwen3-4B-Thinking2026.02 | 80 | 90 | |
| ST-GModel=Qwen3-4B-Thinking2026.02 | 79.37 | 90 | |
| LEDModel=Qwen3-4B-Thinking2026.02 | 78.33 | 90 | |
| CoTModel=Qwen3-4B-Thinking2026.02 | 76.46 | 90 | |
| DoLaModel=Qwen3-4B-Thinking2026.02 | 76.25 | 90 | |
| DoLaModel=MiMo-7B-RL2026.02 | 67.08 | 83.33 | |
| LEDModel=MiMo-7B-RL2026.02 | 66.46 | 83.33 | |
| ST-GModel=MiMo-7B-RL2026.02 | 65.62 | 83.33 | |
| CoTModel=MiMo-7B-RL2026.02 | 65.21 | 83.33 | |
| STModel=MiMo-7B-RL2026.02 | 56.67 | 83.33 |