Multiple-Choice
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
58.5MC1 Accuracy
83
Apr 13, 2026
100Accuracy
42
Jun 5, 2026
73.7Average Accuracy
32
Jun 16, 2026
83.8Accuracy
29
Feb 26, 2026
93.4Accuracy
15
Feb 26, 2026
83COPA Score
14
Apr 21, 2026
93.5Accuracy
11
Mar 17, 2026
67Accuracy
11
Feb 26, 2026
86.7Accuracy
9
Feb 26, 2026
87.7Accuracy
9
Feb 26, 2026
84.4Accuracy
8
Mar 17, 2026
75.5COPA Score
6
May 4, 2026
53.4Ps Score
4
Feb 26, 2026
80.85Accuracy
3
Feb 26, 2026