List your company

Mathematical Reasoning on HMMT-25 (mean@8, pass@8)

50.83Mean@8DeepSeek-R1-Distill-Qwen-32B
10.103620.676831.2541.8232May 29, 2026

Evaluation results

MethodLinks
DeepSeek-R1-Distill-Qwen-32B2026.0550.8370
SkyWork-OR1-Math-7B2026.0532.2255.17
GRPOStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.0532.2255.17
LGRStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.0531.6760
REOPOLDStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.053056.67
OPD (topk)Student Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.0529.5856.67
LGR (topk)Student Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.0527.9253.33
OPD (9k)Student Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.0527.550
JSDStudent Model=DeepSeek-R1-Distill-Qwen-7B, Teacher Model=DeepSeek-R1-Distill-Qwen-32B2026.0525.8343.33
Base ModelStudent Model=DeepSeek-R1-Distill-Qwen-7B2026.0524.1743.33
LGRStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0522.0836.67
LGR (topk)Student Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0522.0846.67
OPD (topk)Student Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0520.527.59
OPD (9k)Student Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0520.4236.67
REOPOLDStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0519.1726.67
JSDStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0518.7523.33
GRPOStudent Model=DeepSeek-R1-Distill-Qwen-1.5B, Teacher Model=SkyWork-OR1-Math-7B2026.0516.7420.69
Base ModelStudent Model=DeepSeek-R1-Distill-Qwen-1.5B2026.0511.6723.33