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

General Multimodal Evaluation on MMMU (val)

69.1Accuracy

GPT-4o

51.9456.39560.8565.305Jul 9, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
69.1
2025.07
67.6
2025.07
66
2025.07
64.7
2025.07
64.4
2025.07
63.1
2025.07
61.6
2025.07
61.4
2025.07
61.1
2025.07
59.3
2025.07
59.1
56
54.1
52.6