Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

MMLU dataset and benchmarks – Page 3 · SOTA2 Research

Research Datasets

MMLU

Benchmarks

Task NameDataset NameSOTA MethodSort

Task NameDataset NameSOTA ResultTrendResults

MMLU-Pro

55Average Score

66

Multi-task Language Understanding

MMLU Pro

96.8Accuracy

66

Language Understanding

MMLU CF

74.5Score

66

Knowledge Evaluation

MMLU

82.03MMLU Accuracy

64

General Knowledge Reasoning

MMLU Pro

82.93Accuracy

64

Multi-task Language Understanding

MMLU-Pro

82.05Accuracy

64

Multi-task Language Understanding

MMLU-Pro

92.85Pass@1

64

General Knowledge Reasoning

MMLU-CF

75.9Accuracy

64

General Reasoning

MMLU-Pro

90.1MMLU-Pro General Reasoning Avg@8 Acc

63

Hallucination Detection

MMLU

87.48AUPRC

62

MMLU-Pro

94.21Average Score

60

Language Understanding

MMLU-Pro

72.8MMLU-Pro Accuracy

60

Knowledge Retention

MMLU (full)

88.3MMLU Accuracy

60

MMLU

0.056Brier Score

58

MMLU

91.35Accuracy

57

Question Answering

MMLU-Redux

92.91Accuracy

57

General Knowledge

MMLU-Pro

82MMLU-Pro General Knowledge Score

56

Question Answering

MMLU-Pro Natural Setting (test)

87.8Accuracy

56

Massive Multitask Language Understanding

MMLU

84.2MMLU Accuracy

55

Question Answering

MMLU

0.141Test Error Probability

52

Multiple Choice Question Answering

MMLU

74.91MMLU Accuracy (Overall)

52

Question Answering

MMLU

64.07AUC

51

General Reasoning

MMLU

86.21Accuracy

51

Question Answering

MMLU

84F1 Score

50

Backdoor Detection

MMLU

0FPR (MMLU)

50

Page 3 of 40

Popular tasks

Language Understanding Multi-task Language Understanding Question Answering Multiple-choice Question Answering Multitask Language Understanding

Follow for updates

Join SOTA2 Contact