AVI-Bench - Visual Sensation QA: leaderboard
Metric: Task score (%) on primitive visual sensation (1,200 multiple-choice items on unfamiliar, low-semantic images and videos): accuracy under double confirmation, 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
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro (Preview 05-06) | 62.67 |
| 2 | Gemini 2.5 Flash (Preview 04-17) | 44.04 |
| 3 | Gemini 2.0 Flash | 40.13 |
| 4 | Phi-4 Multimodal Instruct | 33.26 |
| 5 | Qwen2.5-Omni-7B | 31.51 |
| 6 | Gemini 1.5 Pro | 28.27 |
| 7 | Gemini 1.5 Flash | 24.06 |
Interactive version: theaggregate.ai/benchmark?slug=avi-bench-visual-sensation-qa · How It Works · Data refreshed daily, snapshot 2026-09-29.