Mathematical Problem Solving on AIME 25 (Pass@1, Pass@16)
42.1Pass@1Base Model
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Base ModelModel Scale=Qwen3-4B, Thinking Mode=Thinking2026.04 | 42.1 | 62.3 | |
| MIXED-CUTSModel Scale=Qwen3-4B, Thinking Mode=Non-thinking2026.04 | 41.7 | 71.9 | |
| MIXED-CUTSModel Scale=Qwen3-1.7B, Thinking Mode=Non-thinking2026.04 | 28.1 | 52.5 | |
| GRPOModel Scale=Qwen3-4B, Thinking Mode=Non-thinking2026.04 | 26.6 | 57.9 | |
| Base ModelModel Scale=Qwen3-1.7B, Thinking Mode=Thinking2026.04 | 24.9 | 44.5 | |
| GRPOModel Scale=Qwen3-1.7B, Thinking Mode=Non-thinking2026.04 | 22.8 | 44.5 | |
| Base ModelModel Scale=Qwen3-4B, Thinking Mode=Non-thinking2026.04 | 21.5 | 48 | |
| Base ModelModel Scale=Qwen3-1.7B, Thinking Mode=Non-thinking2026.04 | 11.7 | 27.8 |