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

Spatial Reasoning on CV-Bench

92Accuracy

GEMINI-3-PRO-PREVIEW

56.95266.05175.1584.249Dec 30, 2025Jan 22, 2026Feb 14, 2026Mar 9, 2026Apr 1, 2026Apr 24, 2026May 18, 2026
Updated 14d ago

Evaluation Results

MethodLinks
92
2026.05
87.8
2025.12
86.44
2026.05
86.1
85.9
85.9
2026.05
85.9
2026.05
85.9
2026.05
85.8
2026.01
85.75
2025.12
85.69
2026.05
85.6
2026.02
85.5
85.21
2025.12
85.21
2026.02
85.2
2026.04
85.2
85.1
2026.02
85.1
2026.04
85.1
2026.04
84.9
2026.04
84.7
2026.02
84.6
2026.04
84.6
84.59
2025.12
84.59
2026.02
84.4
2026.05
84.3
84.03
2026.01
83.92
2025.12
83.89
2026.05
83.6
2026.01
82.89
2026.02
82.7
2026.04
82.7
2026.01
82.68
2025.12
82.16
2026.02
82.1
2026.02
81.7
2026.04
81.7
2026.01
81.59
81
2026.02
81
2026.04
81
2026.04
81
2026.05
80.9
2025.12
80.78
2026.05
80.5
78.63
2026.01
78.57
78.43
78.4
2026.05
77.2
76.9
76.5
2026.04
76.5
2026.05
76.5
2026.01
76.22
76
2026.04
76
2025.12
75.8
75.4
2026.05
75.4
75.2
2025.12
75.2
2026.04
75.2
2026.01
74.71
2026.05
74.2
73.7
2026.04
73.7
2026.05
73.7
71.8
2026.02
71.8
2026.04
71.8
2025.12
71.29
2026.05
70.3
2026.05
64.4
2026.01
60.98
2026.05
58.3