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

VLMEvalKit

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal Visual Question AnsweringVLMEvalKit Image Benchmarks
HallBench Accuracy46.8
13
Reasoning and MathVLMEvalKit (test)
MathVista Accuracy73.4
13
Showing 2 of 2 rows