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

Multimodal Generative Tasks on MM-MergeBench (test)

56.65Accuracy

CtM

54.486855.048455.6156.1716Jun 2, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
56.65
2026.06
56.54
2026.06
56.46
2026.06
56.3
2026.06
56.21
2026.06
56.19
2026.06
56.16
2026.06
56.09
2026.06
56.06
2026.06
55.42
2026.06
54.71
2026.06
54.57