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

MMXU

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-image difference analysisMMXU (test)
Worsen Rate23.4
12
Medical Symptom TrackingMMXU
Worsen Rate46.8
7
Multi-image Visual Question AnsweringMMXU (test)
Worsen Rate28.9
3
Showing 3 of 3 rows