List your company

Response Length on MMLU STEM Science Reasoning

1,956Response Length (Tokens)DeepSeek-R1-Distill
446.96838.731,230.51,622.27Jun 16, 2026

Evaluation results

MethodLinks
DeepSeek-R1-DistillModel Scale=1.5B, Base Model=Qwen2.5-Math2026.061,956
DeepSeek-R1-DistillModel Scale=7B, Base Model=Qwen2.5-Math2026.061,572
AdaCoTModel Scale=7B, Base Model=Qwen2.5-Math2026.061,214
AdaCoTModel Scale=1.5B, Base Model=Qwen2.5-Math2026.061,029
AdaptThinkModel Scale=7B, Base Model=Qwen2.5-Math2026.061,023
AdaptThinkModel Scale=1.5B, Base Model=Qwen2.5-Math2026.06995
LHRMsModel Scale=1.5B, Base Model=Qwen2.5-Math2026.06955
S-GRPOModel Scale=1.5B, Base Model=Qwen2.5-Math2026.06812
SuCoModel Scale=1.5B, Base Model=Qwen2.5-Math2026.06745
S-GRPOModel Scale=7B, Base Model=Qwen2.5-Math2026.06717
LHRMsModel Scale=7B, Base Model=Qwen2.5-Math2026.06628
SuCoModel Scale=7B, Base Model=Qwen2.5-Math2026.06505