Skip to content
AI Atlas
BenchmarkActivecategory · knowledge

MMLU

Hendrycks et al.github.com/hendrycks/test

57-subject multiple-choice questions

quality57

Updated 6 h ago · first seen 11 Sept 2026

bench_01M293SPDZHWVW43W1K6V407BF

Metric
accuracy · %
Direction
Results
0
Leader

As of

Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.

Claim history

8 claims · 8 properties

Categorycategory1

Claim history for Category
ValueValid from → toStatusSourceConfidenceExtractor
knowledgecurrentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Creatorcreator1

Claim history for Creator
ValueValid from → toStatusSourceConfidenceExtractor
Hendrycks et al.currentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Known limitationsknown_limitations1

Claim history for Known limitations
ValueValid from → toStatusSourceConfidenceExtractor
Widespread training-data contamination; saturated at the frontier.currentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Metricmetric1

Claim history for Metric
ValueValid from → toStatusSourceConfidenceExtractor
accuracycurrentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Paperpaper1

Claim history for Paper
ValueValid from → toStatusSourceConfidenceExtractor
https://arxiv.org/abs/2009.03300currentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Tasktask1

Claim history for Task
ValueValid from → toStatusSourceConfidenceExtractor
57-subject multiple-choice questionscurrentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Unitunit1

Claim history for Unit
ValueValid from → toStatusSourceConfidenceExtractor
%currentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Websitewebsite1

Claim history for Website
ValueValid from → toStatusSourceConfidenceExtractor
https://github.com/hendrycks/testcurrentcurrentAI Atlas curated registry (YAML, versioned in git, every entry carries its source URL)T2mediumcurated

Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →