MMLU
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MMLU Thai
17Score
3
MMLU financial subsets (test)
38MMLU Business Ethics Acc
3
MMLU-PRO
3ER
3
MMLU-Pro Target Reasked Group (test)
77.78Accuracy (R1)
3
MMLU STEM
60.42Accuracy
3
MMLU STEM
82.3Pass@1
3
MMLU Multilingual (test)
70.9French Accuracy
3
MMLU
0.001MSE
3
MMLU Pro
10.649Acc (Prediction D'_pre)
2
MMLU subjects
99Bracket Score
2
MMLU-Pro
94.4Sel-AUC
2
MMLU
79.8Sycophancy Accuracy (Medicine)
2
MMLU Pro
36.7Accuracy
2
TinyMMLU
66.1E*S Combined Score
2
MMLU
61.2E*S
2
MMLU Pro
66.5Score
2
MMLU STEM (test)
65.2Accuracy
2
MMLU Social Sciences (test)
78.8Accuracy (ACC)
2
MMLU Humanities (test)
59.9Accuracy
2
MMLU
56MMLU Score
2
MMLU-Pro
266Labeling Budget
2
MMLU-Redux
144Labeling Budget
2
MMLU
53.9Accuracy
2
TinyMMLU
32.08Accuracy
2
MMLU ProX-Lite
40.7Accuracy (Afrikaans)
2