List your company

Mathematical Reasoning on IMO-Bench 400 (pass@1, pass@5)

4.88Pass@1DeepSeek-R1-Qwen-1.5B
1.88482.66243.444.2176Jun 29, 2026

Evaluation results

MethodLinks
DeepSeek-R1-Qwen-1.5BGroup Size (G)=162026.064.8814.38
DeepSeek-R1-Qwen-1.5BGroup Size (G)=42026.064.6713.38
DeepSeek-R1-Qwen-1.5BGroup Size (G)=Base2026.063.7510.75
Nemotron-4BGroup Size (G)=42026.062.68.88
Nemotron-4BGroup Size (G)=Base2026.062.359
DeepSeek-7BGroup Size (G)=162026.062.237.88
DeepSeek-7BGroup Size (G)=42026.062.28.25
Nemotron-4BGroup Size (G)=162026.062.179
DeepSeek-7BGroup Size (G)=Base2026.0627.5