Visual Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
100Accuracy (Scale 40)
10
Mar 13, 2026
77Accuracy (Scale 12)
10
Mar 13, 2026
100Accuracy (Scale 8)
10
Mar 13, 2026
56.74Accuracy
10
Mar 13, 2026
59Accuracy
10
Feb 26, 2026
86.5ACC
10
Feb 26, 2026
63.3Accuracy
10
Feb 26, 2026
81.6Accuracy
10
Feb 26, 2026
69.5Accuracy
10
Feb 26, 2026
92.7Color (I) Accuracy
10
Feb 26, 2026
78.23Rel Depth
10
Feb 26, 2026
62.8Accuracy
10
Feb 26, 2026
84.3Object Accuracy
10
Feb 26, 2026
80.11VStar Accuracy
9
Jul 7, 2026
39Score
9
Jun 2, 2026
88.7Pass@1 Score
9
May 28, 2026
44.67Accuracy
9
May 12, 2026
66IoU
9
Apr 28, 2026
69.1Accuracy
9
Mar 4, 2026
38Acc (Largest Lesion Diameter)
9
Feb 26, 2026
60Adjacent Organ Accuracy
9
Feb 26, 2026
98Accuracy
9
Jun 12, 2026
68.5Score
8
Jun 19, 2026
86.7Accuracy
8
Jun 1, 2026
46.34Accuracy
8
Jun 1, 2026