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

MMLUpro

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-task Language UnderstandingMMLUpro (test)
Accuracy95.71
36
Multi-task Complex UnderstandingMMLUPro STEM
Accuracy71.9
9
Showing 2 of 2 rows