OpenEvals - Humanity's Last Exam — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 29 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Kimi K2.5 | 50.2 |
| 2 | DeepSeek V3.2 | 40.8 |
| 3 | GLM-5 | 30.5 |
| 4 | Qwen 3.5 397B A17B | 28.7 |
| 5 | Qwen 3.5 122B A10B | 25.3 |
| 6 | LongCat-Flash-Thinking-2601 | 25.2 |
| 7 | GLM-4.7 | 24.8 |
| 8 | Qwen 3.5 27B | 24.3 |
| 9 | Kimi K2 (Thinking) | 23.9 |
| 10 | Step 3.5 Flash | 23.1 |
| 11 | Qwen 3.5 35B A3B | 22.4 |
| 12 | MiniMax-M2.1 | 22.2 |
| 13 | MiMo-V2-Flash | 22.1 |
| 14 | MiniMax-M2.5 | 19.4 |
| 15 | NVIDIA-Nemotron-3-Super-120B-A12B-BF16 | 18.26 |
Interactive version: theaggregate.ai/benchmark?slug=openevals-humanity-s-last-exam · How the rankings work · Data refreshed daily, snapshot 2026-07-22.