Dynamic-SUPERB: leaderboard
Dynamic speech-language-model benchmark and leaderboard for speech instruction following across many audio tasks.
Metric: Mean Higher-Is-Better Score (%). Source: huggingface.co. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeSTA2.5-Audio | 43.77 | |
| 2 | Qwen2-Audio-7B-Instruct | 37.38 | #880 |
| 3 | Qwen-Audio-Chat | 34.72 | #1244 |
| 4 | Whisper-LLaMA | 30.17 | |
| 5 | SALMONN-7B | 28.14 | #1293 |
| 6 | SALMONN-13B | 27.78 | #1260 |
| 7 | WavLLM | 27.22 | |
| 8 | MU-LLaMA | 21.43 | |
| 9 | LTU-AS | 17.57 | |
| 10 | GAMA-IT | 17.21 |
Interactive version: theaggregate.ai/benchmark?slug=dynamic-superb · How It Works · Data refreshed daily, snapshot 2026-10-20.