Humanity's Last Exam (with tools)
General QA · 2025-01-24
Humanity's Last Exam's 2,500 expert-authored questions answered with external tools available, such as web search, web fetch and code execution. Developers choose and often do not itemize the tools, so setups vary across reports; runs without tools are scored under Humanity's Last Exam.
Top models (higher is better)
| Model | Score |
|---|---|
| Kimi K3.1 Pro | 70.0 |
| Claude Opus 5.5 | 67.7 |
| Claude Fable 5.1 | 65.0 |
| Claude Opus 5 | 64.8 |
| Claude Mythos Preview | 64.7 |
| Claude Fable 5 | 64.5 |
| Claude Sonnet 5.5 | 64.5 |
| DeepSeek V4.1 Flash | 63.9 |
| GLM-5.3 | 62.5 |
| Muse Spark Open | 61.0 |
| DeepSeek V4 Pro 0813 | 60.0 |
| GPT-5.4 Pro | 58.7 |
| GPT-5.6 Sol | 58.0 |
| Opus 4.8 | 57.6 |
| Claude Haiku 5.5 | 57.4 |
| Sonnet 5 | 57.4 |
| GPT-5.5 Pro | 57.2 |
| GPT-6 Astra | 57.2 |
| Kimi K3 | 56.0 |
| Opus 4.7 | 54.7 |
| GLM-5.2 | 54.7 |
| Kimi K2.6 | 54.0 |
| Gemini 3 Deep Think | 53.4 |
| Opus 4.6 | 53.1 |
| GPT-5.5 | 52.2 |
| GPT-5.4 | 52.1 |
| Gemini 3.1 Pro Preview | 51.4 |
| Kimi K2.5 | 50.2 |
| GPT-5.2 Pro | 50.0 |
| Sonnet 4.6 | 49.0 |
| GPT-5.6 Luna | 48.9 |
| Qwen3.5 397B A17B | 48.3 |
| DeepSeek-V4-Pro | 48.2 |
| Inkling-Small | 47.8 |
| Inkling-Small Preview | 46.6 |
| Inkling | 46.0 |
| Gemini 3 Pro Preview | 45.8 |
| GPT-5.2 | 45.5 |
| DeepSeek-V4-Flash | 45.1 |
| Gemini 3 Flash Preview | 43.5 |