Benchmark
Humanity's Last Exam (no tools, text-only)
Scores are percentages.
All reasoning benchmarksScores from LLM Stats data.
- Models measured
- 15
- Scores recorded
- 15
- Sources
- 1
- IQ weight
- 3.04
Score distribution
- 41.7%
- Mean
- 43.4%
- Median
- 12.4%
- Std dev
- 25.8%
- 10th percentile
- 56.9%
- 90th percentile
- 13.3%
- Min
This benchmark has a weight of 3.04 for chat models, measured over 15 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 64.4% |
| 2 | Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 56.9% |
| 3 | GPT-6.1 Sol· LLM Stats data | 52.9% |
| 4 | Hy3 (Deep Infra)· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 47% |
| 5 | ERNIE 5.5 Flash Preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 46.4% |
| 6 | Claude Haiku 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 45.9% |
| 7 | DeepSeek V4.1 Flash 1017· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 44% |
| 8 | Hy4 preview· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43.4% |
| 9 | MiniMax M3.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43.4% |
| 10 | DeepSeek V4 Pro 0813· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 42.7% |
| 11 | DeepSeek V4.1 Flash· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 39.1% |
| 12 | Qwen3 Max· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 30.2% |
| 13 | Inkling FP4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 29.7% |
| 14 | Solar Mini 4· LLM Stats data | 25.8% |
| 15 | Seed 2.0 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 13.3% |
Data credits: Data by LLM Stats