Language Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
48.1Accuracy
10
Apr 23, 2026
26.4Accuracy
10
Apr 23, 2026
55.2Accuracy
10
Mar 18, 2026
55.2Accuracy
10
Mar 18, 2026
55.2Accuracy
10
Mar 18, 2026
25.8TPS
10
Mar 12, 2026
4.051LIS
10
Feb 26, 2026
35.91Dolly MMLU Score
10
Feb 26, 2026
28.7Accuracy
10
Feb 26, 2026
78.12MMLU Utility
9
Jun 16, 2026
6.2MMLU Δ (pts)
8
Jul 7, 2026
26.12HellaSwag Score
8
Jun 15, 2026
89.9Average Accuracy (MMLU)
8
May 7, 2026
47.5Accuracy
8
Apr 6, 2026
0.555HS Accuracy
8
Feb 26, 2026
99.87Language Consistency
8
Feb 26, 2026
49.57ARC (Challenge)
8
Feb 26, 2026
0.445CMMLU
8
Feb 26, 2026
64.1ARC-E Accuracy
8
Feb 26, 2026
49HellaSwag Accuracy
8
Feb 26, 2026
55.6Accuracy
8
Feb 26, 2026
61.39Accuracy
7
Jun 26, 2026
59.01Accuracy
7
Jun 15, 2026
62.9HellaSwag Accuracy
7
May 21, 2026
71.8Health Accuracy
7
Apr 24, 2026