HumanLikeness - Sound-2 — leaderboard
Metric: Humanlike Score (%). Source: huggingface.co. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | starchat2-15B-v0.1 | 70.12 |
| 2 | Phi-3 Mini 4K Instruct | 67.57 |
| 3 | Llama 3.1 8B Instruct | 64.53 |
| 4 | CodeLlama-34B-Instruct-hf | 63.97 |
| 5 | GPT-4o | 62.86 |
| 6 | Mistral Nemo Instruct (2407) | 62.58 |
| 7 | Llama 3.1 70B Instruct | 61.97 |
| 8 | zephyr-7B-alpha | 61.83 |
| 9 | Llama 3 70B Instruct | 61.36 |
| 10 | GPT-3.5 Turbo | 61.25 |
| 11 | Llama 2 7B Chat | 60.54 |
| 12 | c4ai-command-r-plus | 59.93 |
| 13 | GPT-4o Mini | 58.34 |
| 14 | Mistral 7B Instruct (v0.3) | 58.32 |
| 15 | Mistral 7B Instruct (v0.2) | 57.81 |
Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-sound-2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.