OmniMMI Benchmark

What is OmniMMI?

OmniMMI is a cutting-edge benchmark for multi-modal interaction, specifically designed for OmniLLMs in streaming video environments. It includes 1,121 videos…

Released

2025-04-01

Modality

multimodal

Evaluates

多模态, Multimodal, 其他, Other, 多模态模型, VLM, 视频理解, Video Understanding, 跨模态推理, Cross-modal Reasoning, 不支持, Unsupported

Openness

open

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/OmniMMI

OmniMMI paper, code and dataset

• Paper https://arxiv.org/abs/2503.22952
• Code https://github.com/OmniMMI/OmniMMI
• Dataset https://huggingface.co/datasets/ColorfulAI/OmniMMI
• Dataset https://huggingface.co/datasets/bigai-nlco/OmniMMI

Which sources cite OmniMMI?

• OmniMMI

Related 多模态, Multimodal, 其他, Other, 多模态模型, VLM, 视频理解, Video Understanding, 跨模态推理, Cross-modal Reasoning, 不支持, Unsupported benchmarks

• 3DCodeBench
• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• ActiveFly-Bench
• Ada-LEval
• AECBench