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

Expert-level multidisciplinary QA on MMMU (dev val)

52.8Pass@1

Vision-SR1

47.870449.150250.4351.7098Dec 13, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
52.8
2025.12
52.28
2025.12
52.22
2025.12
52.11
2025.12
52.09
2025.12
51.72
2025.12
51.59
2025.12
51.42
2025.12
50.81
2025.12
50.8
2025.12
48.51
2025.12
48.15
2025.12
48.06