Spatial Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
62Accuracy
16
Feb 26, 2026
35.2Maze Navigation Acc
16
Feb 26, 2026
91.3Score
15
May 13, 2026
49.28Accuracy (Pass@1)
15
Apr 10, 2026
82.38Accuracy (Pass@1)
15
Apr 10, 2026
96.312D Relationship Score
15
Mar 17, 2026
86.67Rotation Score
15
Feb 26, 2026
63.91Accuracy
15
Feb 26, 2026
51.9Measurement Accuracy
15
Feb 26, 2026
55.5C-OVO
14
Jun 19, 2026
55.9SPBench Score
14
Apr 7, 2026
54.3LLM-Judge Accuracy
14
Apr 21, 2026
72.5Object Count Score
14
Mar 19, 2026
83.62D Score
14
Feb 26, 2026
70.2SI Score
14
Feb 26, 2026
80.67Accuracy
13
Jun 18, 2026
47.4Score
13
May 13, 2026
72.9Overall Accuracy
13
May 22, 2026
69.2Count Score
13
Apr 24, 2026
88.5Relative Score
13
Apr 24, 2026
41.17Score
13
Apr 14, 2026
72Consistency Score
13
Apr 10, 2026
77.7Original Score
13
Apr 10, 2026
68.6Height Score
13
Apr 10, 2026
67.89SITE-Bench Score
13
Apr 7, 2026