AudioBench - Clotho AQA: leaderboard
AudioBench Clotho AQA slice measuring question answering over spoken or audio content.
Metric: Mean Score (%). Source: github.com. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen-Audio-Chat | 61.93 | #1244 |
| 2 | SALMONN-7B | 57.75 | #1293 |
| 3 | Qwen2-Audio-7B-Instruct | 53.89 | #880 |
| 4 | SeaLLMs-Audio-7B | 53.04 | #1018 |
| 5 | Phi-4 Multimodal Instruct | 47.87 | #906 |
Interactive version: theaggregate.ai/benchmark?slug=audiobench-clotho-aqa · How It Works · Data refreshed daily, snapshot 2026-10-20.