AudioBench - YouTube PQA: leaderboard
AudioBench YouTube PQA slice measuring question answering over spoken or audio content.
Metric: Mean Score (%). Source: github.com. Saturation forecast: Estimated already saturated. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 1.5 Flash | 49.91 | #618 |
| 2 | Qwen-Audio-Chat | 37.16 | #1244 |
| 3 | Qwen2-Audio-7B-Instruct | 36.98 | #880 |
| 4 | SALMONN-7B | 32.12 | #1293 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=audiobench-youtube-pqa · How It Works · Data refreshed daily, snapshot 2026-10-20.