Natural Language Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
72RTE Accuracy
12
Feb 26, 2026
89.5BoolQ
12
Feb 26, 2026
97.9Intent Accuracy
12
Feb 26, 2026
86.33MNLI Accuracy
11
Jul 1, 2026
86.86CoLA Accuracy
11
Mar 4, 2026
71.73CoLA MCC
11
Feb 26, 2026
80.5GLUE Average Score
11
Feb 26, 2026
62.3WG Accuracy
11
Feb 26, 2026
57.1Accuracy
11
Feb 26, 2026
6.2Accuracy
11
Feb 26, 2026
54.5Accuracy
11
Feb 26, 2026
88BoolQ Accuracy
11
Feb 26, 2026
91RTE Accuracy
10
May 12, 2026
88.69RTE Accuracy
10
May 11, 2026
88.97MRPC Score
10
Apr 21, 2026
88.88HA Score
10
Apr 21, 2026
87.63CoLA Accuracy
10
Feb 27, 2026
86.7BoolQ
10
Feb 26, 2026
42.96HellaSwag
10
Feb 26, 2026
91.3Mpqa Accuracy
10
Feb 27, 2026
87.6MNLI-m Accuracy
10
Feb 26, 2026
85.7MRPC Accuracy
9
Jun 4, 2026
48.9DCLM Core Score
9
May 18, 2026
29.8ARC-C Accuracy
9
May 18, 2026
96GPU Utilisation
9
May 13, 2026