Humanity's Last Exam (Text Only): leaderboard
Humanity's Last Exam text-only leaderboard evaluates frontier LLMs using text-based expert questions, excluding multimodal content.
Metric: Score. Source: scale.com. Status: saturation imminent. 61 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) (High) | 47.31 |
| 2 | Claude Fable 5.1 (xHigh) | 46.8 |
| 3 | GPT-5.4 Pro (xHigh) | 45.32 |
| 4 | Muse Spark | 40.92 |
| 5 | Gemini 3 Pro (Preview) | 37.72 |
| 6 | GPT-5.4 (xHigh) | 36.47 |
| 7 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | 36.24 |
| 8 | GPT-5 Pro | 33.32 |
| 9 | GPT-5.2 | 28.5 |
| 10 | GPT-5 | 26.32 |
| 11 | Claude Opus 4.5 (20251101) (Thinking) | 26.32 |
| 12 | GPT-5.1 (Thinking) | 24.65 |
| 13 | Gemini 2.5 Pro (Preview 06-05) | 22.06 |
| 14 | O3 (High) | 20.57 |
| 15 | O3 (Medium) | 19.78 |
Interactive version: theaggregate.ai/benchmark?slug=humanity-s-last-exam-text-only · How It Works · Data refreshed daily, snapshot 2026-09-05.