General Reasoning & Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
82.1Accuracy
21
Apr 14, 2026
72.6Accuracy (RealWorldQA)
21
Apr 14, 2026
82.1Accuracy
21
Apr 14, 2026
66.8Accuracy (MMStar General Reasoning)
14
May 14, 2026
18.9Accuracy
8
Apr 14, 2026
73.6Accuracy
8
Apr 14, 2026
47.3Accuracy
8
Apr 14, 2026