OmniCap-IF - Audio-Visual Content: leaderboard

Metric: Constraint satisfaction rate (%) on joint audio-visual content constraints of multi-constraint captioning instructions on videos with sound; format constraints are checked by rule-based tools on LLM-extracted structure, content constraints by temporal-IoU tools or a gpt-5-mini judge answering binary and multiple-choice questions against the caption; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 14 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)73.4
2Gemini 3 Flash72.35
3MiMo-V2-Omni68.95
4MiMo-V2.567.68
5MiniCPM-o-4.551.86
6Qwen3 Omni 30B A3B Instruct49.31
7Qwen2.5-Omni-7B34.88

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