CASU - Scene Description Event Recall: leaderboard
Metric: Event match (%; an LLM judge counts the share of the scene script's sound events that the model's free description of the clip recognises). Source: arxiv.org. Saturation forecast: Around December 2026. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Flash | 75 |
| 2 | GPT-4o Audio | 72 |
| 3 | Qwen3 Omni 30B A3B Instruct | 65 |
| 4 | Qwen2.5-Omni-7B | 44 |
| 5 | Qwen2-Audio-7B-Instruct | 41 |
| 6 | SALMONN-13B | 36 |
| 7 | Voxtral-Small-24B-2507 | 27 |
Interactive version: theaggregate.ai/benchmark?slug=casu-scene-description-event-recall · How It Works · Data refreshed daily, snapshot 2026-09-26.