Benchmark · Active
Humanity's Last Exam
Created by Center for AI Safety and Scale AI, 23 Jan 2025 · Higher is better · Website
Settings and run setups
| Task set | Tools | Providers |
|---|
| Full set (2,500 questions) | No tools | |
| Full set (2,500 questions) | With tools | |
| Text-only subset | No tools | |
| Text-only subset | With tools | |
| Not stated | No tools | |
| Not stated | With tools | |
| Full set (2,500 questions) | Not stated | |
| Text-only subset | Not stated | |
| Not stated | Not stated | |
Download image
All benchmarks