HumanLikeness - Overall — leaderboard
Psycholinguistic humanlikeness benchmark comparing model response distributions with 2,000 human participants across sound, word, syntax, meaning, and discourse tasks.
Metric: Overall Humanlike (%). Source: huggingface.co. Status: saturation imminent. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 66.91 |
| 2 | Llama 3.1 8B Instruct | 66.2 |
| 3 | Phi-3 Mini 4K Instruct | 64.73 |
| 4 | Mistral Nemo Instruct (2407) | 63.97 |
| 5 | Llama 2 13B Chat Base | 63.26 |
| 6 | CodeLlama-34B-Instruct-hf | 62.16 |
| 7 | c4ai-command-r-plus | 61.09 |
| 8 | Llama 3 8B Instruct | 61.01 |
| 9 | starchat2-15B-v0.1 | 60.84 |
| 10 | GPT-4o | 58.98 |
| 11 | GPT-3.5 Turbo | 58.66 |
| 12 | Yi 1.5 34B Chat | 58.5 |
| 13 | Llama 2 7B Chat | 57.6 |
| 14 | Llama 3 70B Instruct | 57.14 |
| 15 | zephyr-7B-alpha | 56.89 |
Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-overall · How the rankings work · Data refreshed daily, snapshot 2026-07-22.