AV-Odyssey - Scene Recognition: leaderboard

The "Scene Recognition" subtask of AV-Odyssey. A task from AV-Odyssey, an audio-visual benchmark evaluating multimodal models on joint understanding of audio and video content.

Metric: Score (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 18 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 1.5 Pro30#445
2Gemini 1.5 Flash29.5#618
3Gemini 1.5 Flash-8B29#819

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=av-odyssey-scene-recognition · How It Works · Data refreshed daily, snapshot 2026-10-20.