LLM Stats (Humanity's Last Exam (with tools, text-only)): leaderboard
LLM Stats mirror of Humanity’s Last Exam run two ways at once: tools enabled, and the multimodal questions dropped so the set is text alone. Both choices lift the numbers well above the bare no-tools board, so read this column against other tool-enabled runs rather than against plain HLE.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around 2029. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 67.7 | #3 |
| 2 | DeepSeek V4 Pro (0813) | 60 | #53 |
| 3 | GPT-6 Astra | 57.2 | #6 |
| 4 | Qwen 3.8 Max | 56.2 | #35 |
| 5 | Hy4 preview | 55.4 | #38 |
| 6 | Hy3 | 53.2 | #93 |
| 7 | Qwen 3 Max (Thinking) | 49.8 | #208 (Qwen 3 Max) |
| 8 | Step 3.7 Flash | 47.2 | #236 |
| 9 | Inkling | 46 | #126 |
| 10 | Seed 1.8 | 40.9 | #139 |
No result here: #7 GPT-6.1 Sol, #8 Gemini 4 Argon, #9 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-humanity-s-last-exam-with-tools-text-only · How It Works · Data refreshed daily, snapshot 2026-10-20.