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

CMath

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningCMATH
Accuracy95.7
50
Mathematical ReasoningCMATH (test)
Accuracy89.7
25
Mathematical ReasoningCMath
Pass@191.7
19
MathCMATH
Score96.9
10
Chinese-language abilityCMATH
Accuracy84.8
6
Mathematical Problem SolvingCMATH
Accuracy71.7
4
Chinese Mathematical ReasoningCMath
CMath Score40.5
1
Showing 7 of 7 rows