Humanity's Last Exam
Expert-level questions
knowledge and reasoningactive
Definition
- Version
- not in the notes
- Category
- knowledge and reasoning
- Maintainer
- not in the notes
- Status
- active
- Home page
- https://lastexam.ai/
Sources
- Best AI models, October 2026: Opus 5.5, GPT-6, Gemini slash-digital.io https://slash-digital.io/en/insights/ai-models-2026/
- Humanity's Last Exam lastexam.ai https://lastexam.ai/
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
 Claude Opus 5.5 | 67.7 % | vendor-reported | with tools (Anthropic) | – | – | slash-digital.io |
| Claude Opus 5.5 | 61.4 % | independent | AA-measured | – | – | slash-digital.io |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Humanity's Last Exam", https://aurion.semai.id/benchmarks/humanitys-last-exam/, records of October 2026.