AA Humanity's Last Exam: leaderboard
Artificial Analysis independent evaluation of HLE: 2,500 expert-vetted frontier questions across math, sciences, and humanities.
Metric: Accuracy (%). Source: artificialanalysis.ai. Saturation forecast: Around September 2027. 668 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 57.09 | #8 (Gemini 4 Argon) |
| 2 | Claude Opus 5 (Max) | 54.87 | #12 (Claude Opus 5) |
| 3 | GPT-6 Astra (Max) | 54.68 | #6 (GPT-6 Astra) |
| 4 | GPT-6 Astra (xHigh) | 54.59 | #6 (GPT-6 Astra) |
| 5 | Claude Opus 5 (xHigh) | 54.4 | #12 (Claude Opus 5) |
| 6 | Muse Wonder Full (Max) | 53.61 | #10 (Muse Wonder Full) |
| 7 | GPT-6 Astra (High) | 53.06 | #6 (GPT-6 Astra) |
| 8 | GPT-6.1 Sol (Max) | 52.92 | #7 (GPT-6.1 Sol) |
| 9 | Claude Opus 5 (High) | 52.83 | #12 (Claude Opus 5) |
| 10 | GPT-6 Astra (Medium) | 52.73 | #6 (GPT-6 Astra) |
| 11 | GPT-6.1 Sol (xHigh) | 52.64 | #7 (GPT-6.1 Sol) |
| 12 | GPT-6.1 Sol (High) | 51.39 | #7 (GPT-6.1 Sol) |
| 13 | Claude Opus 5 (Medium) | 51.3 | #12 (Claude Opus 5) |
| 14 | Muse Wonder (Max) | 50.09 | #15 (Muse Wonder) |
| 15 | GPT-6.1 Sol (Medium) | 49.86 | #7 (GPT-6.1 Sol) |
No result here: #3 Claude Opus 5.5, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=aa-humanity-s-last-exam · How It Works · Data refreshed daily, snapshot 2026-10-20.