HumanLikeness - Meaning-2: leaderboard
HumanLikeness meaning task comparing model responses with human semantic illusion behavior.
Metric: Humanlike Score (%). Source: huggingface.co. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Nemo Instruct (2407) | 75.1 | #936 |
| 2 | Llama 3.1 8B Instruct | 73.74 | #1035 |
| 3 | zephyr-7B-alpha | 73.13 | #1312 |
| 4 | zephyr-7B-beta | 72.93 | #1330 |
| 5 | Llama 2 13B Chat Base | 71.65 | #1422 |
| 6 | Llama 3.1 70B Instruct | 66.52 | #561 |
| 7 | Llama 3 8B Instruct | 65.66 | #1129 |
| 8 | Yi 1.5 34B Chat | 64.86 | #800 |
| 9 | Mistral 7B Instruct (v0.2) | 64.02 | #1333 |
| 10 | CodeLlama-34B-Instruct-hf | 63.11 | #1381 |
| 11 | GPT-4o | 60.82 | #343 |
| 12 | Llama 2 7B Chat | 60.09 | #1499 |
| 13 | c4ai-command-r-plus | 59.23 | #724 |
| 14 | Mixtral 8x7B Instruct (v0.1) | 59.03 | #1073 |
| 15 | Llama 3 70B Instruct | 58.91 | #634 |
Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-meaning-2 · How It Works · Data refreshed daily, snapshot 2026-10-20.