Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AMC23 (Response Accuracy %)
Loading...
90.5
Response Accuracy
S-GRPO
21.236
39.218
57.2
75.182
Jun 16, 2026
Response Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Response Accuracy
S-GRPO
Model Scale=7B
2026.06
90.5
SuCo
Model Scale=7B
2026.06
90.3
LHRMs
Model Scale=7B
2026.06
87.3
AdaptThink
Model Scale=7B
2026.06
82
AdaCoT
Model Scale=7B
2026.06
81.5
DeepSeek-R1-Distill
Model Scale=7B
2026.06
75.5
SuCo
Model Scale=1.5B
2026.06
73.8
LHRMs
Model Scale=1.5B
2026.06
70
S-GRPO
Model Scale=1.5B
2026.06
69.5
AdaptThink
Model Scale=1.5B
2026.06
65.8
AdaCoT
Model Scale=1.5B
2026.06
62.5
DeepSeek-R1-Distill
Model Scale=1.5B
2026.06
56.5
Math-Instruct
Model Scale=7B
2026.06
53.5
Math-Instruct
Model Scale=1.5B
2026.06
43.5
Math-Base
Model Scale=7B
2026.06
36.2
Math-Base
Model Scale=1.5B
2026.06
23.9
Feedback
Search any
task
Search any
task