Mathematical Reasoning on AIME 25 (Avg, Pass)
71.1Avg Score (AIME 25)DistIL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DistILModel=Qwen3-8B2026.06 | 71.1 | 85 | |
| OPSDModel=Qwen3-8B2026.06 | 69.7 | 83.3 | |
| SDPOModel=Qwen3-8B2026.06 | 68.3 | 84 | |
| BaseModel=Qwen3-8B2026.06 | 66.9 | 79.6 | |
| GRPOModel=Qwen3-8B2026.06 | 66.9 | 79.6 | |
| SFTModel=Qwen3-8B2026.06 | 65.3 | 80.7 | |
| DistILModel=Qwen3-4B2026.06 | 55.3 | 77.6 | |
| OPSDModel=Qwen3-4B2026.06 | 51.5 | 76.1 | |
| BaseModel=Qwen3-4B2026.06 | 50.3 | 69.5 | |
| GRPOModel=Qwen3-4B2026.06 | 50.3 | 69.5 | |
| SDPOModel=Qwen3-4B2026.06 | 49.6 | 74.1 | |
| SFTModel=Qwen3-4B2026.06 | 43.1 | 67.4 |