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

Video Reasoning on Video-MMMU

84.6Accuracy

GPT-5

16.68834.31951.9569.581May 11, 2025Jul 5, 2025Aug 29, 2025Oct 23, 2025Dec 17, 2025Feb 10, 2026Apr 6, 2026
Updated 11d ago

Evaluation Results

MethodLinks
2026.01
84.6
2026.01
83.6
2025.05
81.4
2025.05
76.7
2025.05
72.1
2026.01
61.2
2026.01
61.2
2026.04
61.2
2026.04
54.2
2026.01
53.9
2026.01
53.4
2026.01
53.1
2026.01
52.7
2026.01
52.6
2026.01
52.3
2026.04
52.3
2026.01
51.6
51.4
2026.01
51.3
2026.01
51.3
2026.01
51.1
2026.04
51.1
2026.04
50.2
2026.01
50.1
50
2026.01
49.8
2026.01
49.4
2026.01
49.4
2026.04
49.2
2026.01
49
2026.01
47.4
2026.01
47.4
47.4
46
2026.01
45.8
2026.01
40.8
36.1
2026.01
35.2
2026.01
33.8
2026.01
33.8
33.8
2026.01
32
2026.01
23.2
20.9
2026.01
19.3