Vellum - Humanity's Last Exam: leaderboard
Vellum's independently-run Humanity's Last Exam evaluation. Expert-level questions across academic domains.
Metric: Accuracy (%). Source: www.vellum.ai. Saturation forecast: Around February 2027. 41 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 67.7 | #3 |
| 2 | Claude Fable 5.1 | 65 | #9 |
| 3 | Claude Mythos 5.1 | 65 | #1 |
| 4 | Claude Opus 5 | 64.7 | #12 |
| 5 | Claude Sonnet 5.5 | 64.5 | #11 |
| 6 | Claude Mythos 5 | 64.5 | #2 |
| 7 | Claude Opus 4.8 | 57.9 | #37 |
| 8 | Claude Sonnet 5 | 57.4 | #57 |
| 9 | GPT-6 Astra | 57.2 | #6 |
| 10 | Kimi K3 | 56 | #30 |
| 11 | GLM-5.3 Flash | 55.3 | #65 |
| 12 | Gemini 3.8 Flash | 54.9 | #21 |
| 13 | GLM-5.2 | 54.7 | #73 |
| 14 | Kimi K2.6 | 54 | #104 |
| 15 | GPT-5.4 | 52.1 | #79 |
No result here: #7 GPT-6.1 Sol, #8 Gemini 4 Argon, #10 Muse Wonder Full.
Interactive version: theaggregate.ai/benchmark?slug=vellum-humanity-s-last-exam · How It Works · Data refreshed daily, snapshot 2026-10-20.