MMLU
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MMLU
79.64MMLU Accuracy
41
MMLU Biz
90F1 Score
41
MMLU Pro
76.88MMLU Pro Engineering Acc
41
MMLU on-the-fly
0.524BRR
40
MMLU-Pro Phys.
70.4Accuracy (%)
40
MMLU-Pro Law
30.8Accuracy
40
MMLU
69MMLU Score
40
MMLU-R
84.4Accuracy (MMLU-R General Reasoning)
40
MMLU
89.08MMLU Accuracy
39
MMLU
73.59MMLU Score
39
MMLU
82.4Accuracy
39
MMLU Pro
86.7pass@1
38
M-MMLU (test)
59.6Overall Accuracy
38
MMLU PRO
78MMLU-Pro Reasoning Score
36
MMLU (test)
92.87Accuracy (Natural Sciences)
36
MMLU-Pro
95.7Accuracy
36
MMLU Medical subjects
80Anatomy (EN) Accuracy
35
MMLU-Pro Health (EN) X (test)
68.85Accuracy (%)
35
MMLU Professional Medecine FR (test)
84.19Accuracy
35
MMLU Professional Medecine EN (test)
85.29Accuracy
35
MMLU Medical Genetics FR (test)
86Accuracy
35
MMLU Medical Genetics EN (test)
94Accuracy (MMLU Medical Genetics)
35
MMLU Medical Subjects (test)
92.36Accuracy (College Biology, EN)
35
MMLU
72.28EM
35
MMLU
82Accuracy
35