README.md

---
license: apache-2.0
language:
- en
base_model:
- tencent/HunyuanVideo
- Qwen/Qwen2.5-VL-7B-Instruct
pipeline_tag: any-to-any
tags:
- video
---
<div align="center">

<h1><b>UniVideo: Unified Understanding, Generation, and Editing for Videos</b></h1>

</div>

<div align="center">
  
**[Cong Wei<sup>*,1,2</sup>](https://congwei1230.github.io/) &ensp;**
**[Quande Liu<sup>†,2</sup>](https://liuquande.github.io/) &ensp;**
**[Zixuan Ye<sup>2</sup>](https://openreview.net/profile?id=~Zixuan_Ye1) &ensp;** 
**[Qiulin Wang<sup>2</sup>](https://scholar.google.com/citations?user=3vvZdy8AAAAJ&hl=en) &ensp;**
**[Xintao Wang<sup>2</sup>](https://xinntao.github.io/)**

**[Pengfei Wan<sup>2</sup>](https://magicwpf.github.io/) &ensp;**
**[Kun Gai<sup>2</sup>](https://openreview.net/profile?id=~Kun_Gai1) &ensp;**
**[Wenhu Chen<sup>†,1</sup>](https://wenhuchen.github.io/)**
  <p>
    <sup>1</sup>University of Waterloo &nbsp;&nbsp;
    <sup>2</sup>Kling Team, Kuaishou Technology<br>
    <sup>*</sup>Work done during an internship at Kling Team, Kuaishou Technology
    <sup>†</sup>Corresponding author
  </p>
</div>

<p align="center">
  <a href='https://congwei1230.github.io/UniVideo/'><img src='https://img.shields.io/badge/Project-Page-Green'></a>
  &nbsp;
  <a href="https://arxiv.org/abs/2510.08377"><img src="https://img.shields.io/static/v1?label=Arxiv&message=UniVideo&color=red&logo=arxiv"></a>
  &nbsp;
  <a href='https://huggingface.co/KlingTeam/UniVideo'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Model-orange'></a>
  &nbsp;
  <a href="https://github.com/KlingTeam/UniVideo">
    <img src="https://img.shields.io/badge/GitHub-Code-red?logo=github">
  </a>
</p>


## 🔔News

- [2026-01-07]: Released [Code](https://github.com/KlingTeam/UniVideo) and [Model](https://huggingface.co/KlingTeam/UniVideo).
- [2025-10-09]: Released [Arxiv Preprint](https://arxiv.org/abs/2510.08377) and the [Project Page](https://congwei1230.github.io/UniVideo/)

## How to use

- Please refer to [**🔗 GitHub**](https://github.com/KlingTeam/UniVideo) for usage.


## Acknowledgement

- [HunyuanVideo](https://github.com/Tencent-Hunyuan/HunyuanVideo): the base video generation model used in this work. Thanks to the authors for their excellent contribution.
- [Qwen2.5-VL](https://github.com/QwenLM): the base vlm model used in this work. Thanks to the authors for their excellent contribution.
- [MetaQueries](https://xichenpan.com/metaquery/): we adopt their query implementation. Thanks to the authors for their excellent contribution.

## 🌟 Citation

If you find UniVideo useful for your research and applications, please cite using this BibTeX:

```bibtex
@article{wei2025univideo,
  title={Univideo: Unified understanding, generation, and editing for videos},
  author={Wei, Cong and Liu, Quande and Ye, Zixuan and Wang, Qiulin and Wang, Xintao and Wan, Pengfei and Gai, Kun and Chen, Wenhu},
  journal={arXiv preprint arXiv:2510.08377},
  year={2025}
}
```