ModelRefs / Humanity's Last Exam Leaderboard
Humanity's Last Exam Leaderboard
Multi-disciplinary expert-level exam, contamination resistant. Current leaders, methodology, and citation sources for Humanity's Last Exam.
Overview
Multi-disciplinary expert-level exam, contamination resistant.
How it is measured: Closed-book; >3k questions across 100+ domains.
How this benchmark is scored
| Category | reasoning |
|---|---|
| Maximum score | 100 % accuracy |
| Direction | Higher is better |
Primary source: https://lastexam.ai/
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Humanity's Last Exam Leaderboard.
Frequently asked questions
What does the Humanity's Last Exam benchmark measure?
Multi-disciplinary expert-level exam, contamination resistant.
What is the methodology?
Closed-book; >3k questions across 100+ domains.
Which AI model is the current Humanity's Last Exam leader?
Gemini 3 Pro (Google) currently leads with 37.5 % accuracy.