HumanLikeness - Sound-1: leaderboard
HumanLikeness sound-association task comparing model responses with human sound-shape association patterns.
Metric: Humanlike Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 20 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.1 70B Instruct | 88.51 | #561 |
| 2 | c4ai-command-r-plus | 79.19 | #724 |
| 3 | CodeLlama-34B-Instruct-hf | 78.53 | #1381 |
| 4 | Mistral Nemo Instruct (2407) | 74.09 | #936 |
| 5 | Llama 2 7B Chat | 73.76 | #1499 |
| 6 | Llama 3.1 8B Instruct | 73.14 | #1035 |
| 7 | Llama 3 8B Instruct | 68.62 | #1129 |
| 8 | Yi 1.5 34B Chat | 66.54 | #800 |
| 9 | Phi-3 Mini 4K Instruct | 60.76 | #1297 |
| 10 | Mixtral 8x7B Instruct (v0.1) | 59.96 | #1073 |
| 11 | Llama 3 70B Instruct | 59.94 | #634 |
| 12 | GPT-4o | 59.56 | #343 |
| 13 | starchat2-15B-v0.1 | 58.11 | #832 |
| 14 | Llama 2 13B Chat Base | 56.71 | #1422 |
| 15 | zephyr-7B-alpha | 56.6 | #1312 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=humanlikeness-sound-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.