Humanity's Last Exam (Text Only)
General QA · 2025-01-24
This Humanity's Last Exam subtrack covers expert-authored questions that use text rather than image inputs. Scores report answer accuracy in percentage points.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 3.1 Pro Preview | 47.3 |
| GPT-5.4 Pro | 45.3 |
| Muse Spark | 40.9 |
| Gemini 3 Pro Preview | 37.7 |
| GPT-5.4 | 36.5 |
| Opus 4.6 | 36.2 |
| GPT-5 Pro | 33.3 |
| GPT-5.2 | 28.5 |
| Opus 4.5 | 26.3 |
| GPT-5 | 26.3 |
| GPT-5.1 | 24.6 |
| Gemini 2.5 Pro Preview 06-05 | 22.1 |
| o3 | 20.6 |
| GPT-5 Mini | 19.7 |
| O4 Mini | 18.9 |
| Gemini 2.5 Pro Preview 05-06 | 18.4 |
| Gemini 2.5 Pro Preview 03-25 | 18.4 |
| gpt-oss-120b | 15.5 |
| Qwen3 235B A22B Thinking 2507 | 15.4 |
| Sonnet 4.5 | 14.1 |
| DeepSeek-R1-0528 | 14.0 |
| o3-mini | 13.4 |
| DeepSeek-V3.1 | 12.9 |
| Gemini 2.5 Flash Preview 04-17 | 12.6 |
| Qwen3-235B-A22B | 11.8 |
| Opus 4.1 | 11.3 |
| Opus 4 | 10.8 |
| Gemini 2.5 Flash Preview 05-20 | 10.7 |
| gpt-oss-20b | 9.7 |
| GLM-4.5 | 9.6 |
| GLM 4.5 Air | 9.4 |
| DeepSeek-R1 | 8.5 |
| Gemini 3.1 Flash-Lite Preview | 8.0 |
| Claude 3.7 Sonnet | 7.9 |
| O1 | 7.8 |
| O1 Pro | 7.7 |
| Sonnet 4 | 7.6 |
| Gemini 2.0 Flash Thinking Experimental 01-21 | 6.5 |
| GPT-5.1 Instant | 6.5 |
| GPT-4.5 | 5.8 |