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

Multimodal Reasoning on MMStar (Accuracy)

77.1Accuracy

LaRe

6.27624.66343.0561.437Sep 25, 2025Nov 4, 2025Dec 14, 2025Jan 24, 2026Mar 5, 2026Apr 14, 2026May 25, 2026
Updated 7d ago

Evaluation Results

MethodLinks
2025.11
77.1
2025.11
75.3
75.2
74.7
2025.11
74
2026.04
72.63
2025.11
72.6
2026.04
72.27
2026.04
72.2
2026.04
72.2
2026.04
72
71.83
2026.04
70.3
2026.05
70
2026.05
69.9
2026.04
69.47
69.3
2026.05
68.9
2026.05
68.3
2026.05
68
2026.05
67.3
2026.05
67.2
2025.11
67.1
2026.05
66.6
2026.04
65.9
2025.11
65.9
2026.05
65.8
2026.05
65.3
2026.05
65.1
2025.09
63.9
2026.05
63.6
2026.04
63.2
2025.09
63.2
62.9
2025.09
62.8
2026.05
62.7
2026.05
62.5
2026.04
62
2026.05
61.7
2026.05
61.6
2026.05
61.3
2026.04
60.47
2026.04
60.33
2026.04
60.27
2026.04
59.7
2026.05
59.5
2026.05
59.2
2026.04
59.13
2026.05
59
2026.05
58.9
2026.04
58.73
2026.05
58.2
58.1
2026.04
57.93
2026.05
57.6
2026.05
57.1
2026.05
57.1
2026.05
56.8
2026.05
56.2
2026.05
54
2025.10
54
2025.10
54
2026.05
53.5
2026.04
53.33
2026.05
52.1
2026.05
51.6
2025.10
47
2025.09
43.6
41
2025.10
32
2025.10
29
2025.10
29
2025.10
14
2025.10
14
2025.10
12
2025.10
11
2025.10
11
2025.10
9