Mathematical Reasoning on MetaMathQA 2000
86.93Pass@1DeepSeek-R1-Qwen-1.5B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DeepSeek-R1-Qwen-1.5BGroup Size (G)=162026.06 | 86.93 | 96.53 | |
| DeepSeek-R1-Qwen-1.5BGroup Size (G)=42026.06 | 85.01 | 96.45 | |
| DeepSeek-R1-Qwen-1.5BGroup Size (G)=Base2026.06 | 82.71 | 95.95 | |
| DeepSeek-7BGroup Size (G)=162026.06 | 73.07 | 91.78 | |
| DeepSeek-7BGroup Size (G)=42026.06 | 70.3 | 90.88 | |
| DeepSeek-7BGroup Size (G)=Base2026.06 | 66.93 | 90.65 | |
| Nemotron-4BGroup Size (G)=162026.06 | 52.55 | 75.43 | |
| Nemotron-4BGroup Size (G)=42026.06 | 49.73 | 75.15 | |
| Nemotron-4BGroup Size (G)=Base2026.06 | 46.54 | 74.05 |