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

Multi-modal Video Understanding on MVBench

82.9Accuracy

Gemini-2.5-Pro (minimal)

-3.31619.06741.4563.833May 31, 2025Jul 26, 2025Sep 20, 2025Nov 16, 2025Jan 11, 2026Mar 8, 2026May 4, 2026
Updated 22d ago

Evaluation Results

MethodLinks
82.9-----------------------
2025.05
77.1-72.1---------------------
2025.05
76.1-64.6---------------------
2026.05
75.6-----------------------
2026.05
75.4-----------------------
2025.05
74.4-67---------------------
2026.03
74.3-----------------------
73.8-----------------------
2025.05
73.6-62.7---------------------
2025.12
72.3-----------------------
72.3-----------------------
72.1-----------------------
2026.03
71.8-----------------------
2026.05
71.8-----------------------
2025.05
71.5-60.8---------------------
2026.05
71.5-----------------------
2025.12
71.4-----------------------
71.1-----------------------
2026.03
70.9-----------------------
2026.03
70.9-----------------------
2026.03
70.7-----------------------
70.6-----------------------
70.5-----------------------
2025.12
70.4-----------------------
2025.12
70.4-----------------------
2026.03
69.9-----------------------
2026.05
69.9-----------------------
69.8-----------------------
2025.12
69.6-----------------------
2026.03
69.6-----------------------
2026.05
69.6-----------------------
69.5-----------------------
2025.12
69.2-----------------------
2025.12
68.9-----------------------
2026.01
68.8-----------------------
2026.03
68.7-----------------------
2026.03
68.6-----------------------
68.4-----------------------
2025.12
67.9-----------------------
2026.01
67-----------------------
2025.12
67-----------------------
2025.05
67-58---------------------
2025.12
64.6-----------------------
2026.02
64.6-----------------------
2026.03
64.3-----------------------
2026.02
63.9-----------------------
2025.12
61.2-----------------------
2025.12
60.5-----------------------
2025.05
59.4-58---------------------
2026.02
59-----------------------
2026.03
58.4-----------------------
2026.04
58.4----------------------100
2026.04
57.9----------------------96.7
2026.03
57.5-----------------------
2026.05
57-----------------------
2026.01
56.7-----------------------
2026.03
56.7-----------------------
2025.05
56.7-53.2---------------------
2026.05
56.7-----------------------
2026.01
56.5-----------------------
2026.03
56.5-----------------------
2026.02
55.8-----------------------
2026.03
55.8-----------------------
2026.03
55.1-----------------------
2026.04
54.8----------------------95
2026.02
54.6-----------------------
2026.04
54.2----------------------88.5
2026.04
54.2----------------------93.8
2026.04
53.7----------------------86.5
2026.03
53.4-----------------------
2026.04
53.2----------------------91
2025.05
53.1-50.2---------------------
2026.02
52.9-----------------------
2026.04
52.4----------------------89.4
2026.03
51.1-----------------------
2026.03
49.6-----------------------
2026.04
48.1----------------------82.1
2026.03
46.5-----------------------
2026.03
45.5-----------------------
2025.05
43.7-----------------------
2026.03
37-----------------------
2025.12
33.9-----------------------
2026.04
0----------------------12.6
2026.02
--34.1513422.525.527.5403730212932.532.522.522.54840.5384345.539-
2026.02
--35.556352726.533.5413623.523.533.526.542.520.525.55340.53048.54640.5-
2026.02
--38.5613426.53640.53635293532.536.543.52019.553.541.52966.54253-
2026.02
--37.159.52838.533.534.53122.5274929.53836.52624.5513332.552.537.560-
2026.02
--4273503933.535.5393232.55130.539.53733.526484034.567.540.560-
2026.02
-58.3----------------------
2026.02
-56.4----------------------
2026.02
-57.8----------------------
2026.02
-56.8----------------------
2026.02
-58.1----------------------
2026.02
-58----------------------
2026.02
-56----------------------
2026.02
-57.6----------------------
2026.02
-56.2----------------------
2026.02
-58.1----------------------
2026.02
-58.2----------------------
2026.02
-55----------------------
Showing 100 of 189 rows