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

Multi-modal Video Evaluation on Video-MME Long v1.0 (test)

57.89Score

Full Cache

46.096449.158252.2255.2818Apr 17, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
57.8968.95-
2026.04
56.4563.267.73
2026.04
56.4563.267.73
2026.04
55.2270.99-
2026.04
54.5659.9611.03
2026.04
53.8967.46-
2026.04
53.1166.680.78
2026.04
52.7860.018.84
2026.04
52.3366.50.96
2026.04
52.3366.50.96
2026.04
52.2262.764.7
2026.04
5265.232.23
2026.04
51.3364.033.43
2026.04
5065.52-
2026.04
49.5654.4611.06
2026.04
49.3364.471.05
2026.04
49.1161.314.21
2026.04
48.7764.281.24
2026.04
4860.624.9
2026.04
46.5555.829.7