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

Visual Question Answering on MMMU

81.7Accuracy

Gemini 2.5 Pro

29.658443.169256.6870.1908Mar 23, 2026Mar 25, 2026Mar 28, 2026Mar 31, 2026Apr 3, 2026Apr 6, 2026Apr 9, 2026
Updated 9d ago

Evaluation Results

MethodLinks
81.7
2026.03
76.4
76
2026.04
71.6
2026.03
70.8
2026.04
70.7
2026.04
70.6
70.1
2026.04
69.8
2026.04
69.7
2026.04
67.9
2026.04
60.2
2026.04
58.8
2026.04
57.3
2026.04
56.7
2026.04
55.8
2026.04
53.4
2026.03
36.61
2026.03
36.4
2026.03
36.38
2026.03
35.82
2026.03
35.8
2026.03
35.72
2026.03
35.67
2026.03
35.52
2026.03
35.14
2026.03
35.05
2026.03
34.68
2026.03
34.59
2026.03
34.4
2026.03
34.27
2026.03
33.75
2026.03
33.01
2026.03
32.79
2026.03
32.15
2026.03
31.9
2026.03
31.66