HumanLikeness - Word-1 — leaderboard

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

Top models

#ModelScore
1GPT-4o70.88
2GPT-3.5 Turbo68.9
3c4ai-command-r-plus65.8
4GPT-4o Mini64.81
5Llama 3.1 70B Instruct63.71
6CodeLlama-34B-Instruct-hf62.43
7Llama 3.1 8B Instruct62.18
8Phi-3 Mini 4K Instruct61.24
9Llama 2 7B Chat60.24
10starchat2-15B-v0.160.05
11Llama 3 8B Instruct59.96
12Mistral Nemo Instruct (2407)58.69
13Llama 3 70B Instruct57.26
14Yi 1.5 34B Chat57.21
15Mistral 7B Instruct (v0.2)56.32

Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-word-1 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.