HumanLikeness - Meaning-2 — leaderboard

Metric: Humanlike Score (%). Source: huggingface.co. 20 models tracked.

Top models

#ModelScore
1Mistral Nemo Instruct (2407)75.1
2Llama 3.1 8B Instruct73.74
3zephyr-7B-alpha73.13
4zephyr-7B-beta72.93
5Llama 2 13B Chat Base71.65
6Llama 3.1 70B Instruct66.52
7Llama 3 8B Instruct65.66
8Yi 1.5 34B Chat64.86
9Mistral 7B Instruct (v0.2)64.02
10CodeLlama-34B-Instruct-hf63.11
11GPT-4o60.82
12Llama 2 7B Chat60.09
13c4ai-command-r-plus59.23
14Mixtral 8x7B Instruct (v0.1)59.03
15Llama 3 70B Instruct58.91

Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-meaning-2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.