HumanLikeness - Discourse-2 — leaderboard
Metric: Humanlike Score (%). Source: huggingface.co. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 13B Chat Base | 78.85 |
| 2 | Phi-3 Mini 4K Instruct | 76.18 |
| 3 | zephyr-7B-alpha | 74.58 |
| 4 | Mistral Nemo Instruct (2407) | 67.84 |
| 5 | Llama 3.1 70B Instruct | 63.12 |
| 6 | GPT-4o | 62.11 |
| 7 | starchat2-15B-v0.1 | 61.81 |
| 8 | CodeLlama-34B-Instruct-hf | 60.74 |
| 9 | zephyr-7B-beta | 59.56 |
| 10 | GPT-4o Mini | 57.32 |
| 11 | Llama 3.1 8B Instruct | 55.57 |
| 12 | Llama 3 8B Instruct | 54.28 |
| 13 | Mixtral 8x7B Instruct (v0.1) | 52.18 |
| 14 | c4ai-command-r-plus | 50.44 |
| 15 | Llama 3 70B Instruct | 50.4 |
Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-discourse-2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.