Multiple-Choice Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
66.29Accuracy
20
Mar 25, 2026
39.16Accuracy
17
May 14, 2026
65.7Accuracy
11
Mar 9, 2026
32.8Accuracy
10
Apr 20, 2026
78.2Accuracy
8
Feb 26, 2026
48.39Accuracy
6
May 14, 2026
32.64Accuracy
4
Mar 23, 2026
78Acc (CN)
4
Feb 26, 2026
80.6Final Test Mean Score@16
3
Jun 4, 2026