OmniCap-IF - Visual Content: leaderboard

Metric: Constraint satisfaction rate (%) on 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)74.15
2Gemini 3 Flash73.6
3MiMo-V2-Omni70.14
4MiMo-V2.569.82
5MiniCPM-o-4.559.24
6Qwen3 Omni 30B A3B Instruct58.13
7Qwen2.5-Omni-7B47.68

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