Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

MATH500

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningMath500 Level 5
Accuracy71.1
54
Advanced Mathematical ReasoningMath500 512 tokens
Pass@1 Accuracy46.3
15
Advanced Mathematical ReasoningMath500 256 tokens
Pass@1 Accuracy41.2
15
ReasoningMATH500
MATH500 Average Score96.4
12
Mathematical ReasoningMATH500
Pass@185.6
10
Open Question AnsweringMATH500 (test)
Accuracy0.94
9
Question AnsweringMath500 (in-domain)
Accuracy95.2
5
Showing 7 of 7 rows