AVI-Bench - Audio-Visual Localization: leaderboard

Metric: Task score (%) on audio-visual localization (205 image and audio pairs): 0.7 times the mean IoU of the boxes predicted for the sounding objects plus 0.3 times an instance-count score, on AVI-Bench (5,864 audio-visual items in 14 tasks over four cognitive stages), zero-shot with the same prompts for every omni-modal model; task scores normalized to percentages; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 26 models tracked.

Top models

#ModelScore
1Gemini 1.5 Pro43.63
2Gemini 2.5 Flash (Preview 04-17)39.18
3Gemini 2.5 Pro (Preview 05-06)39.13
4Gemini 2.0 Flash37.93
5Gemini 1.5 Flash36.33
6Qwen2.5-Omni-7B19.36
7Phi-4 Multimodal Instruct10.61

Interactive version: theaggregate.ai/benchmark?slug=avi-bench-audio-visual-localization · How It Works · Data refreshed daily, snapshot 2026-09-29.