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

Multi-modal Vision-Language Evaluation on MMVet

46.8Accuracy

VisionZip

20.69627.47334.2541.027Feb 5, 2026Feb 21, 2026Mar 9, 2026Mar 25, 2026Apr 10, 2026Apr 26, 2026May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.02
46.8
2026.02
46.1
2026.02
45
2026.02
44.7
2026.02
44.1
2026.02
43.8
2026.02
42.3
2026.02
42.3
2026.02
42.1
2026.02
41.7
2026.02
41.4
2026.02
41.3
2026.02
41.1
2026.02
40.7
2026.02
39.7
2026.02
39.2
2026.02
39
2026.05
38.8
2026.05
38.2
2026.05
37.6
2026.02
36.6
2026.05
34.4
2026.05
33.6
2026.05
33.4
2026.05
33.4
2026.05
32.8
2026.05
32.5
2026.05
32.4
2026.05
32
2026.05
31.5
2026.05
31.1
2026.02
30.2
2026.05
27.5
2026.05
26.7
2026.05
26.1
2026.05
25.9
2026.05
24.6
2026.05
21.7