# Humanity's Last Exam (no tools) — AI model rankings

Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better.

48 tracked models have a published Humanity's Last Exam (no tools) score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Claude Opus 5.5 | Anthropic | 64.4% | Lab | Sep 22 2026 |
| 2 | Claude Fable 5.1 | Anthropic | 60.9% | Lab | Sep 1 2026 |
| 3 | Claude Mythos 5 | Anthropic | 59% | Lab | Jun 9 2026 |
| 4 | Claude Fable 5 | Anthropic | 57.8% | Lab | Jun 9 2026 |
| 5 | Claude Sonnet 5.5 | Anthropic | 56.9% | Lab | Sep 28 2026 |
| 6 | Claude Mythos Preview | Anthropic | 56.8% | Lab | Apr 7 2026 |
| 7 | Claude Opus 5 | Anthropic | 56.3% | Lab | Jul 24 2026 |
| 8 | Kimi K3.1 Pro | Moonshot AI | 51.1% | Lab | Oct 23 2026 |
| 9 | Claude Opus 4.8 | Anthropic | 49.8% | Lab | May 28 2026 |
| 10 | Claude Opus 4.7 | Anthropic | 46.9% | Lab | Apr 16 2026 |
| 11 | Claude Haiku 5.5 | Anthropic | 45.9% | Lab | Oct 7 2026 |
| 12 | Gemini 3.1 Pro | Google | 44.4% | Lab | Feb 19 2026 |
| 13 | Kimi K3 | Moonshot AI | 43.5% | Lab | Jul 16 2026 |
| 14 | Claude Sonnet 5 | Anthropic | 43.2% | Lab | Jun 30 2026 |
| 15 | GPT-5.5-Pro | OpenAI | 43.1% | Lab | Apr 23 2026 |
| 16 | GPT-5.4-Pro | OpenAI | 42.7% | Lab | Mar 5 2026 |
| 16 | DeepSeek-V4-Pro-0813 | DeepSeek | 42.7% | Lab | Aug 13 2026 |
| 18 | DeepSeek-V4.1-Flash-1017 | DeepSeek | 42.6% | Lab | Oct 17 2026 |
| 19 | GPT-5.5 | OpenAI | 41.4% | Lab | Apr 23 2026 |
| 20 | GLM-5.2 | Z.ai | 40.5% | Lab | Jun 16 2026 |
| 21 | Gemini 3.5 Flash | Google | 40.2% | Lab | May 19 2026 |
| 22 | Claude Opus 4.6 | Anthropic | 40% | Lab | Feb 5 2026 |
| 23 | GPT-5.4 | OpenAI | 39.8% | Lab | Mar 5 2026 |
| 24 | Gemini 3.0 Pro | Google | 37.5% | Lab | Nov 18 2025 |
| 25 | DeepSeek-V4.1-Flash | DeepSeek | 36.8% | Lab | Sep 10 2026 |
| 26 | Qwen3.8-Flash-Next | Qwen | 35.9% | Lab | Aug 26 2026 |
| 27 | GPT-5.2 | OpenAI | 34.5% | Lab | Dec 11 2025 |
| 28 | Gemini 3.0 Flash | Google | 33.7% | Lab | Dec 17 2025 |
| 29 | Claude Sonnet 4.6 | Anthropic | 33.2% | Lab | Feb 17 2026 |
| 30 | Claude Opus 4.5 | Anthropic | 30.8% | Lab | Nov 24 2025 |
| 30 | Qwen3.8-27B | Qwen | 30.8% | Lab | Aug 14 2026 |
| 32 | Qwen3.5 | Qwen | 28.7% | Lab | Feb 16 2026 |
| 33 | GPT-5.4 mini | OpenAI | 28.2% | Lab | Mar 17 2026 |
| 34 | GPT-5 | OpenAI | 24.8% | Lab | Aug 7 2025 |
| 34 | GLM-4.7 | Z.ai | 24.8% | Lab | Dec 22 2025 |
| 36 | GPT-5.4 nano | OpenAI | 24.3% | Lab | Mar 17 2026 |
| 37 | Qwen3.6 | Qwen | 21.4% | Lab | Apr 16 2026 |
| 38 | Gemma 4 | Google | 19.5% | Lab | Apr 2 2026 |
| 39 | Gemini 2.5 Pro | Google | 18.8% | Lab | Mar 25 2025 |
| 40 | Claude Sonnet 4.5 | Anthropic | 17.7% | Lab | Sep 29 2025 |
| 41 | GPT-5 mini | OpenAI | 16.7% | Lab | Aug 7 2025 |
| 42 | Gemini 3.1 Flash-Lite | Google | 16% | Lab | Mar 3 2026 |
| 43 | gpt-oss-120b | OpenAI | 14.9% | Lab | Aug 5 2025 |
| 44 | Gemini 2.5 Flash | Google | 12.1% | Lab | Apr 17 2025 |
| 45 | gpt-oss-20b | OpenAI | 10.9% | Lab | Aug 5 2025 |
| 46 | Claude Haiku 4.5 | Anthropic | 10.2% | Lab | Oct 15 2025 |
| 47 | GPT-5 nano | OpenAI | 8.7% | Lab | Aug 7 2025 |
| 48 | Gemini 2.5 Flash-Lite | Google | 5.1% | Lab | Jun 17 2025 |


---

Canonical page: https://aireleasetracker.com/benchmark/hle-no-tools
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.