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

General Visual Reasoning on MME-RealWorld-Lite

73.06Accuracy

GPT4o

37.23246.533555.83565.1365Feb 9, 2026Feb 26, 2026Mar 15, 2026Apr 1, 2026Apr 18, 2026May 5, 2026May 22, 2026
Updated 9d ago

Evaluation Results

MethodLinks
2026.02
73.06
2026.04
67.1
2026.02
62.85
2026.04
60.8
2026.04
59.3
2026.04
58.3
2026.02
57.63
2026.04
57.22
2026.02
56.8
2026.04
56.5
2026.04
56.48
2026.02
56.23
2026.02
55.34
2026.04
55.2
2026.02
54.56
2026.02
51.69
2026.04
50.2
2026.02
50.03
2026.05
48.8
2026.04
48.6
2026.02
48.57
2026.02
48.31
2026.04
47.94
2026.05
46.7
2026.02
46.43
2026.04
46.2
2026.05
46.1
2026.02
45.34
44.9
2026.02
44.87
2026.04
44.1
2026.02
43.72
43.7
42.3
2026.02
42
2026.04
40.02
2026.02
38.61