General Reasoning on MATH (pass@1)
89.6pass@1STAR-1
Evaluation Results
| Method | Links | |
|---|---|---|
| STAR-1Model Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 89.6 | |
| SInternalModel Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 89.6 | |
| BaseModel Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 89.2 | |
| STAR-1Model Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 88.8 | |
| SInternalModel Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 88.4 | |
| BaseModel Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 88.2 | |
| BaseModel Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 85.2 | |
| SafeChainModel Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 85.2 | |
| SInternalModel Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 84.4 | |
| SafeChainModel Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 82.4 | |
| STAR-1Model Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 81.8 | |
| SafeChainModel Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 74.2 |