Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Mathematical Reasoning on AMC (Accuracy, Token count)

97.5Accuracy

Vanilla

68.976.32583.7591.175Apr 8, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.04
97.56,562
2026.04
97.55,035
2026.04
97.54,514
2026.04
97.53,975
2026.04
956,807
2026.04
954,894
2026.04
954,617
2026.04
92.57,232
2026.04
92.54,966
2026.04
905,448
2026.04
904,507
2026.04
87.55,155
2026.04
77.52,065
2026.04
72.52,264
2026.04
701,738