Knowledge & Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
71.58Exact Match
20
Apr 21, 2026
93.3MMLU Redux Score
15
May 13, 2026
56.22K&R Score (NPM)
14
Mar 5, 2026
79.8Accuracy
9
Feb 26, 2026
74.4Accuracy (MedXpertQA Text+MM)
6
Apr 8, 2026
81Score
5
Mar 27, 2026
87.7Accuracy
4
Feb 26, 2026
21.2Exact Match
4
Feb 26, 2026
77.9EM
4
Feb 26, 2026
63.4Accuracy
4
Feb 26, 2026