General Language Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
77.3Overall Accuracy
14
Feb 26, 2026
1.015Average Normalized Performance
13
Jun 16, 2026
85.93Overall Accuracy
13
Mar 13, 2026
4.88PPL
13
Feb 26, 2026
71.6Accuracy
12
Apr 10, 2026
59.4ARC-c Accuracy (25-shot)
12
Mar 19, 2026
71.09Average Accuracy
11
May 28, 2026
99.7Sports Score
11
Feb 26, 2026
99.4Sports Score
11
Feb 26, 2026
80.83RTE Accuracy
11
Feb 26, 2026
75.7Average Accuracy
10
Jul 2, 2026
0.559MMLU Score
10
Feb 26, 2026
55.55ARC Score
9
Apr 24, 2026
62.03ARC
9
Feb 26, 2026
82.6Winogrande Accuracy
8
May 28, 2026
0.492Avg Accuracy
8
Feb 26, 2026
62.6LiDiRus (Corr)
8
Feb 26, 2026
68.84BBH
7
May 11, 2026
65.51Average Score
7
Feb 26, 2026
56.82MMLU-Pro Score
6
Mar 10, 2026
5.13PPL
6
Feb 26, 2026
72.01COLA Score
6
Feb 26, 2026
73.5Utility Score
5
Jun 9, 2026
76.89Accuracy
5
Apr 8, 2026
21.23TPS
4
Jun 4, 2026