Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Response Length on MMLU STEM Science Reasoning
Loading...
1,956
Response Length (Tokens)
DeepSeek-R1-Distill
446.96
838.73
1,230.5
1,622.27
Jun 16, 2026
Response Length (Tokens)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Response Length (Tokens)
DeepSeek-R1-Distill
Model Scale=1.5B, Base...
2026.06
1,956
DeepSeek-R1-Distill
Model Scale=7B, Base M...
2026.06
1,572
AdaCoT
Model Scale=7B, Base M...
2026.06
1,214
AdaCoT
Model Scale=1.5B, Base...
2026.06
1,029
AdaptThink
Model Scale=7B, Base M...
2026.06
1,023
AdaptThink
Model Scale=1.5B, Base...
2026.06
995
LHRMs
Model Scale=1.5B, Base...
2026.06
955
S-GRPO
Model Scale=1.5B, Base...
2026.06
812
SuCo
Model Scale=1.5B, Base...
2026.06
745
S-GRPO
Model Scale=7B, Base M...
2026.06
717
LHRMs
Model Scale=7B, Base M...
2026.06
628
SuCo
Model Scale=7B, Base M...
2026.06
505
Feedback
Search any
task
Search any
task