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

Multimodal Video Evaluation on Video-MME Sub (test)

87.8Accuracy

Gemini 2.5 Pro

58.26465.93273.681.268Mar 30, 2026
Updated 19d ago

Evaluation Results

MethodLinks
2026.03
87.8
2026.03
87.5
2026.03
86.9
2026.03
84.2
2026.03
82.3
2026.03
80.5
2026.03
76.9
2026.03
76.2
2026.03
75.8
2026.03
75.7
2026.03
75.7
2026.03
75.6
2026.03
75.2
2026.03
74
2026.03
73.5
2026.03
69.7
2026.03
69.7
2026.03
69.2
2026.03
68.6
2026.03
68.6
2026.03
65.4
2026.03
59.4