Benchmark RadarAI BENCHMARK PROFILE
AVI-Bench
AVI-Bench evaluates omni-multimodal LLMs on audio-visual tasks across perception, understanding, and reasoning stages, with an extension probing primitive audio-visual sensation using unfamiliar low-semantic stimuli.
- Released
- 2026-06-01
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
Provides a structured way to measure joint audio-visual capabilities in multimodal models, which is currently under-evaluated.
Motivation
Recent advances in Omni-Multimodal Large Language Models (Omni-MLLMs) have enabled strong integration of vision, audio, and language.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.