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

Visual Perception and Reasoning on V* Bench

98.3Attribute Score

Human

22.27642.01361.7581.487Mar 14, 2026Mar 19, 2026Mar 24, 2026Mar 29, 2026Apr 3, 2026Apr 8, 2026Apr 13, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.04
98.310099
2026.04
95.793.494.8
2026.04
94.892.193.7
2026.04
94.892.193.7
2026.04
93.990.892.7
2026.04
93.992.193.2
2026.04
93.992.193.2
2026.04
9390.892.1
2026.04
9392.192.6
2026.04
9390.892.1
2026.04
9389.591.6
2026.04
92.290.891.6
2026.04
92.290.891.6
2026.04
92.289.591.1
2026.04
92.290.891.6
2026.04
92.290.891.6
2026.03
92.1786.8490.05
2026.04
91.388.290.1
2026.04
91.389.590.6
2026.04
91.389.590.6
2026.04
91.389.590.6
2026.04
89.289.589.3
2026.03
88.786.8487.96
2026.03
86.9686.8486.91
2026.03
86.097581.68
2026.03
86.0973.4781.17
2026.03
86.0978.9583.25
2026.03
84.3580.2682.72
2026.04
83.580.382.2
83.576.380.6
2026.03
83.4882.8983.25
2026.03
81.7490.7985.34
2026.03
81.7492.1185.86
2026.03
81.7473.6878.35
2026.03
74.7882.8978.01
2026.04
72.260.567.5
2026.04
51.360.555
2026.04
43.556.648.7
40.959.248.2
2026.04
31.346.137.2
25.247.434