List your company
ResearchFormerly WizWand

Science Reasoning on GPQA Diamond (Response Length (Tokens))

6,858Response Length (Tokens)DeepSeek-R1-Distill
1,170.242,646.874,123.55,600.13Jun 16, 2026
Updated 4mo ago

Evaluation results

MethodLinks
DeepSeek-R1-DistillModel Scale=7B, Base Model=Qwen2.5-Math2026.066,858
DeepSeek-R1-DistillModel Scale=1.5B, Base Model=Qwen2.5-Math2026.066,582
AdaCoTModel Scale=7B, Base Model=Qwen2.5-Math2026.063,993
AdaptThinkModel Scale=7B, Base Model=Qwen2.5-Math2026.063,677
AdaCoTModel Scale=1.5B, Base Model=Qwen2.5-Math2026.063,279
AdaptThinkModel Scale=1.5B, Base Model=Qwen2.5-Math2026.062,914
S-GRPOModel Scale=1.5B, Base Model=Qwen2.5-Math2026.062,828
S-GRPOModel Scale=7B, Base Model=Qwen2.5-Math2026.062,453
LHRMsModel Scale=1.5B, Base Model=Qwen2.5-Math2026.062,381
LHRMsModel Scale=7B, Base Model=Qwen2.5-Math2026.062,042
SuCoModel Scale=1.5B, Base Model=Qwen2.5-Math2026.061,550
SuCoModel Scale=7B, Base Model=Qwen2.5-Math2026.061,389