General Reasoning on AIME 2024 (pass@16)
86.7pass@16Base
Evaluation results
| Method | Links | |
|---|---|---|
| BaseModel Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 86.7 | |
| SInternalModel Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 86.7 | |
| SafeChainModel Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 83.3 | |
| STAR-1Model Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 83.3 | |
| SInternalModel Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 83.3 | |
| BaseModel Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 83.3 | |
| SafeChainModel Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 83.3 | |
| STAR-1Model Architecture=DeepSeek-R1-Distill-Qwen-14B2026.05 | 83.3 | |
| BaseModel Architecture=DeepSeek-R1-Distill-Qwen-7B2026.05 | 80 | |
| SInternalModel Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 80 | |
| STAR-1Model Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 76.7 | |
| SafeChainModel Architecture=DeepSeek-R1-Distill-Llama-8B2026.05 | 66.7 |