Mathematical Reasoning on Math-500 (pass@1, pass@5)
75.74Pass@1DeepSeek-R1-Qwen-1.5B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DeepSeek-R1-Qwen-1.5BGroup Size (G)=162026.06 | 75.74 | 89.6 | |
| DeepSeek-R1-Qwen-1.5BGroup Size (G)=42026.06 | 73.18 | 88.7 | |
| DeepSeek-R1-Qwen-1.5BGroup Size (G)=Base2026.06 | 70.8 | 85.2 | |
| DeepSeek-7BGroup Size (G)=162026.06 | 16.12 | 33.5 | |
| DeepSeek-7BGroup Size (G)=42026.06 | 15.48 | 33.3 | |
| DeepSeek-7BGroup Size (G)=Base2026.06 | 15 | 32.4 | |
| Nemotron-4BGroup Size (G)=162026.06 | 13.34 | 31.5 | |
| Nemotron-4BGroup Size (G)=42026.06 | 12.62 | 31.6 | |
| Nemotron-4BGroup Size (G)=Base2026.06 | 11.64 | 31.4 |