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

Spatial Reasoning on CV-Bench 2D

94Accuracy

Qwen2.5-VL-3B

40.533654.414368.29582.1757Dec 7, 2025Dec 22, 2025Jan 6, 2026Jan 21, 2026Feb 5, 2026Feb 20, 2026Mar 8, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.03
94
2026.03
85.1
2026.03
84.7
2026.03
83.3
2026.03
80
2026.03
68.8
2026.03
68.8
2026.03
68.6
2026.03
66.1
2025.12
55.58
2025.12
55.07
2025.12
54.28
2025.12
54.06
2025.12
52.87
2025.12
52.56
2025.12
48.16
2025.12
47.47
2025.12
46.77
2025.12
44.44
2025.12
43.64
2026.03
43.1
2025.12
42.59