Multiple-choice Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
24.3MMLU Accuracy
10
May 14, 2026
61.43Acc (Norm)
10
Apr 23, 2026
30.43Normalized Accuracy
10
Apr 23, 2026
27.97Normalized Accuracy
10
Apr 23, 2026
45.73Normalized Accuracy
10
Apr 23, 2026
51.94Acc (Normalized)
10
Apr 23, 2026
28.98Acc (Norm)
10
Apr 23, 2026
79Accuracy
10
Apr 3, 2026
83.2Accuracy
10
Mar 27, 2026
47.9Accuracy
10
Feb 26, 2026
67.61Accuracy
10
Feb 26, 2026
56.84Accuracy
10
Feb 26, 2026
72.49Accuracy
10
Feb 26, 2026
71.1Accuracy
10
Feb 26, 2026
9.38Mean EMR
10
Feb 26, 2026
0.037Brier Score
9
Jun 11, 2026
84.2Exact Match Accuracy
9
May 15, 2026
79.7MCQ Exact Accuracy
9
May 15, 2026
30.81MMLU Accuracy (zero-shot)
9
Apr 27, 2026
46.89Accuracy (maj@2)
9
Apr 23, 2026
36.8Accuracy (ZH)
9
Apr 21, 2026
84.5CT WG1 Accuracy
9
Apr 20, 2026
62MMLU Pro (NoMath) Score
9
Apr 10, 2026
124.35AAC
9
Apr 3, 2026
85.07Exact Match (EM)
9
Mar 11, 2026