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

Multimodal Understanding on MMMU

81.8Accuracy

GPT-5

53.7261.0168.375.59Apr 9, 2024Aug 7, 2024Dec 5, 2024Apr 5, 2025Aug 3, 2025Dec 1, 2025Apr 1, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2026.02
81.8-
2026.02
76-
2026.03
74.8-
74.7-
2026.03
73.6-
2026.03
73.4-
2026.02
71.4-
2026.03
71.2-
2026.02
70.8-
2026.02
70.7-
2026.03
70.6-
70.1-
70.1-
2026.03
70-
2024.12
69.9-
2024.12
69.3-
69.1-
68.3-
2025.06
68.1-
2026.03
66.8-
65.9-
2025.06
65.6-
2025.06
64.6-
64.5-
2024.12
64.3-
62.7-
62.2-
2026.03
62.2-
2024.12
61.7-
2024.12
61.178.1
2025.12
60.9-
2024.12
60.8-
2025.02
60.7-
60.6-
60.3-
2025.06
60.3-
2025.09
60.2-
2025.06
59.7-
2024.12
59.677.5
2024.12
59.677.6
2026.03
59.2-
2025.09
59.1-
2026.04
58.994.2
2025.12
58.7-
2026.04
58.7100
2026.01
58.6-
2026.03
58.6-
2026.03
58.6-
2026.04
58.694
2026.03
58.4-
2024.12
58.3-
2026.02
58.3-
2024.10
58.2-
2024.12
58.173.2
58.1-
2026.02
58-
2026.02
58-
2026.02
58-
2026.03
58-
2024.10
57.95-
2026.04
57.886.2
2024.12
57.775
2024.12
57.775.8
2026.04
57.789.6
2025.09
57.6-
2025.12
57.4-
2026.03
57.4-
2026.04
57.493.4
2024.12
57.376.6
2026.02
57.2-
2026.02
57.1-
2026.02
57.1-
2026.04
57.190.8
2026.02
56.9-
2025.10
56.86-
2024.04
56.8-
2024.08
56.8-
2025.12
56.8-
2025.12
56.8-
56.8-
2026.04
56.875.1
2026.03
56.7-
2024.12
56.6-
2026.02
56.5-
2024.10
56.48-
2026.04
56.487.2
2025.12
56.2-
2024.12
56.174.3
2026.04
56.181.3
2024.12
55.972.2
2025.10
55.88-
2024.12
55.670.8
2026.04
55.684
2026.03
55.3-
2025.02
55.2-
2026.01
55.2-
2024.12
55.1-
2026.01
55-
2024.12
54.870.7
2026.04
54.878.9
Showing 100 of 451 rows