Mathematical Reasoning on AIME 25 (avg@256)
12Average Score @256Qwen2.5-7B + DAPO-Silhouette
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2.5-7B + DAPO-SilhouetteModel Family=Qwen Family, Strategy=DAPO-Silhouette2026.01 | 12 | |
| Qwen2.5-7B + DAPOModel Family=Qwen Family, Strategy=DAPO2026.01 | 11.8 | |
| OctoThinker-8B + DAPO-SilhouetteModel Family=OctoThinker Family, Strategy=DAPO-Silhouette2026.01 | 5 | |
| Qwen2.5-7BModel Family=Qwen Family2026.01 | 2.6 | |
| OctoThinker-8B + DAPOModel Family=OctoThinker Family, Strategy=DAPO2026.01 | 2.1 | |
| OctoThinker-8BModel Family=OctoThinker Family2026.01 | 0.7 | |
| Llama-3.1-8B-Instruct + DAPOModel Family=Llama Family, Strategy=DAPO2026.01 | 0.5 | |
| Llama-3.1-8B-InstructModel Family=Llama Family2026.01 | 0.4 | |
| Llama-3.1-8B-Instruct + DAPO-SilhouetteModel Family=Llama Family, Strategy=DAPO-Silhouette2026.01 | 0.4 |