Video spatial reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
79.2Average Score
75
Jun 23, 2026
92.1Configuration Score
20
Mar 9, 2026
81.3Accuracy
16
May 27, 2026
41.4Accuracy
15
May 27, 2026
75Score @ 10 min
5
Jun 19, 2026