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

Ape

Benchmarks

Task NameDataset NameSOTA ResultTrend
Text Math ReasoningAPE210k
Score92.5
12
Quantity TaggingApe clean (test)
Accuracy91.5
4
Target PredictionApe clean (test)
MSE0.44
4
Operation PredictionApe-clean (test)
Accuracy87
4
Common Attribute ComparisonApe-clean (test)
Accuracy87
4
Attribute PredictionApe clean (test)
Accuracy0.86
4
Number Type GroundingApe clean (test)
Accuracy92
4
Number CountingApe clean (test)
MSE0.67
4
Showing 8 of 8 rows