Multitask Language Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
91.5Accuracy
568
Jun 9, 2026
92.16Accuracy
322
Jun 4, 2026
89.31Accuracy
303
Jun 23, 2026
86.3Accuracy
263
Jun 29, 2026
74.12Accuracy
94
May 29, 2026
90.1MDR
45
Apr 15, 2026
84.13Accuracy
42
Jun 12, 2026
92.1Accuracy
42
Feb 26, 2026
79.64MMLU Accuracy
41
Jul 7, 2026
86.7pass@1
38
May 12, 2026
78.3Accuracy
38
Feb 26, 2026
84.19Accuracy
35
Apr 10, 2026
85.29Accuracy
35
Apr 10, 2026
86Accuracy
35
Apr 10, 2026
94Accuracy (MMLU Medical Genetics)
35
Apr 10, 2026
73.3Accuracy
32
Jun 25, 2026
72.8MMLU (CoT)
29
Mar 4, 2026
53.94MMLU Score
24
Jun 18, 2026
71.9pass@1
24
Feb 26, 2026
30.75Acc (Normalized)
22
Apr 23, 2026
90.1Accuracy
21
Feb 26, 2026
65.7vbal Score
18
Apr 7, 2026
99.5Accuracy (Exact)
18
Feb 26, 2026
33.38Accuracy
18
Feb 26, 2026
41.94Accuracy
18
Feb 26, 2026