Mathematical Reasoning on AIME24 (Avg, Pass)
76.9Average ScoreSDPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SDPOModel=Qwen3-8B2026.06 | 76.9 | 89.5 | |
| OPSDModel=Qwen3-8B2026.06 | 76.5 | 91.3 | |
| DistILModel=Qwen3-8B2026.06 | 76.4 | 90.7 | |
| BaseModel=Qwen3-8B2026.06 | 75.9 | 86.8 | |
| GRPOModel=Qwen3-8B2026.06 | 75.9 | 86.8 | |
| SFTModel=Qwen3-8B2026.06 | 74.2 | 85.3 | |
| DistILModel=Qwen3-4B2026.06 | 65.3 | 87.5 | |
| OPSDModel=Qwen3-4B2026.06 | 63.2 | 85.4 | |
| BaseModel=Qwen3-4B2026.06 | 61.4 | 82 | |
| GRPOModel=Qwen3-4B2026.06 | 61.4 | 82 | |
| SDPOModel=Qwen3-4B2026.06 | 60.9 | 85.4 | |
| SFTModel=Qwen3-4B2026.06 | 55.8 | 80.9 |