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

MathBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Math ReasoningMathBench EN
Score50.6
32
Math ReasoningMathBench-CN
General Score48
14
Bilingual Mathematical ReasoningMathBench EN
Accuracy40.1
13
Bilingual Mathematical ReasoningMathBench CN
Accuracy46.2
13
Mathematical ReasoningMathBench
Accuracy95.6
7
Mathematical ReasoningMathBench College
Accuracy78
3
Mathematical ReasoningMathBench High
Accuracy84
3
Mathematical ReasoningMathBench Middle
Accuracy84.67
3
Mathematical ReasoningMathBench Arithmetic
Accuracy75
3
Showing 9 of 9 rows