MMLU
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MMLU-pro
59.33RA
16
MMLU-pro
14,946.31Average Context Length (tokens)
16
MMLU i.d.
87.68Accuracy
15
MMLU-s
76.03P@1
15
MMLU
41.2MMLU Accuracy (ar)
15
MMLU
96Algebra Score
15
MMLU OpenOrca
44.52MMLU OpenOrca Score
15
MMLU Wizard
45.02Score
15
MMLU-Pro
68.21Avg@k
15
MMLU-Pro
75.6Accuracy
15
MMLU STEM
73.7Accuracy (STEM)
15
MMLU-ProX T5
47.1Accuracy
15
MMLU (test)
0Student Information Leakage
15
MMLU v1 (test)
66.6Accuracy
15
MMLU
0.841AUC
15
MMLU
45.2MMLU Math Score
15
MMLU Redux
93.3MMLU Redux Score
15
Global MMLU-Lite
85.72Overall Score
15
MMLU
0.13Mean Per-Step Regret
15
MMLU Bengali (test)
62MMLU Score
15
MMLU
81.31Accuracy
15
MMLU-Pro
60.11Accuracy
15
MMLU-redux
86.8Accuracy
15
MMLU v1 (test)
72.4Accuracy
15
MMLU French (test)
71Accuracy
15