# 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.

14 tracked models have a published Humanity's Last Exam (no tools) score. Higher is better. Scores are as published at each model's release.

## Ranking

| Rank | Model | Developer | Score | Released |
| --- | --- | --- | --- | --- |
| 1 | Claude Opus 5 | Anthropic | 56.3% | Jul 24 2026 |
| 2 | Claude Opus 4.8 | Anthropic | 49.8% | May 28 2026 |
| 3 | Claude Opus 4.7 | Anthropic | 46.9% | Apr 16 2026 |
| 4 | Gemini 3.1 Pro | Google | 44.4% | Feb 19 2026 |
| 5 | Kimi K3 | Moonshot AI | 43.5% | Jul 16 2026 |
| 6 | Claude Sonnet 5 | Anthropic | 43.2% | Jun 30 2026 |
| 7 | GPT-5.5 | OpenAI | 41.4% | Apr 23 2026 |
| 8 | GLM-5.2 | Z.ai | 40.5% | Jun 16 2026 |
| 9 | Gemini 3.5 Flash | Google | 40.2% | May 19 2026 |
| 10 | Gemini 3.0 Flash | Google | 33.7% | Dec 17 2025 |
| 11 | Claude Sonnet 4.6 | Anthropic | 33.2% | Feb 17 2026 |
| 12 | Qwen3.5 | Qwen | 28.7% | Feb 16 2026 |
| 13 | GLM-4.7 | Z.ai | 24.8% | Dec 22 2025 |
| 14 | Qwen3.6 | Qwen | 21.4% | Apr 16 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/hle-no-tools
Full dataset: https://aireleasetracker.com/llms-full.txt · JSON: https://aireleasetracker.com/models.json
Source: AI Release Tracker (https://aireleasetracker.com). Benchmark scores are the figures published by the releasing lab at launch.