---
title: lynx
canonical_url: "https://www.modelscope.cn/models/bytedance-community/lynx"
md_url: "https://www.modelscope.cn/models/bytedance-community/lynx.md"
repository: bytedance-community/lynx
last_updated: 2026-06-05
license: apache-2.0
pipeline_tag: image-to-video
tasks:
  - image-to-video
base_model:
  - Wan-AI/Wan2.1-T2V-14B
base_model_relation: finetune
parameters: 4.8B
tensor_type:
  - F32
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 26
stars: 0
tags:
  - image-to-video
  - video-generation
  - personalized-video
  - identity-preservation
---

# lynx

> lynx - bytedance-community 在 ModelScope 开源的模型。Lynx: High-Fidelity Personalized Video Generation

bytedance-community/lynx 是 ModelScope 魔搭社区上的 4.8B 参数image-to-video模型，采用 apache-2.0 许可，基于 Wan-AI/Wan2.1-T2V-14B 构建。

- **Repository**: bytedance-community/lynx
- **License**: apache-2.0
- **Tasks**: image-to-video
- **Parameters**: 4.8B
- **Base model**: Wan-AI/Wan2.1-T2V-14B
- **Tags**: image-to-video, video-generation, personalized-video, identity-preservation
- **Downloads**: 26
- **Stars**: 0
- **Last updated**: 2026-06-05

Source: https://www.modelscope.cn/models/bytedance-community/lynx

---

<div align="center">

# Lynx: High-Fidelity Personalized Video Generation

<a href="https://arxiv.org/abs/2509.15496"><img src="https://img.shields.io/badge/arXiv-2509.15496-b31b1b.svg"/></a>
<a href="https://byteaigc.github.io/Lynx/"><img src="https://img.shields.io/badge/Project-Page-green.svg"/></a>
<a href="https://github.com/bytedance/lynx"><img src="https://img.shields.io/badge/GitHub-Repo-blue.svg"/></a>

<h5 style="font-size:1.1em; letter-spacing:0.5px;">
<a href="https://ssangx.github.io/">Shen Sang*</a>&nbsp;&nbsp;&nbsp;
<a href="https://tiancheng-zhi.github.io/">Tiancheng Zhi*</a>&nbsp;&nbsp;&nbsp;
<a href="https://gutianpei.github.io/">Tianpei Gu</a>&nbsp;&nbsp;&nbsp;
<a href="https://www.jingliu.net/">Jing Liu</a>&nbsp;&nbsp;&nbsp;
<a href="https://linjieluo.github.io/">Linjie Luo</a>
</h5>

<p style="font-size: 1.05em; margin: 8px 0;">
Intelligent Creation, ByteDance
</p>

<p style="font-size: 0.95em; font-style: italic;">
* Equal Contribution
</p>

<br>
<img src="assets/teaser.jpg" width="400"/>

</div>

Lynx is a state-of-the-art high-fidelity personalized video generation model that creates videos from a single input image while preserving the subject's identity. Built on a Diffusion Transformer (DiT) foundation model with lightweight ID-adapters and Ref-adapters for identity preservation and spatial detail enhancement.

## Model Variants

This repository contains two model variants:

- **Lynx Full Model** (`lynx_full`): Complete version with all advanced features and best performance
- **Lynx Lite Model** (`lynx_lite`): Lightweight model with fewer parameters (no Ref-adapter), tailored for efficient 24fps (121-frame) video generation.

## Citation

If you use this model in your research, please cite:

```bibtex
@article{sang2025lynx,
  title={Lynx: Towards High-Fidelity Personalized Video Generation},
  author={Sang, Shen and Zhi, Tiancheng and Gu, Tianpei and Liu, Jing and Luo, Linjie},
  journal={arXiv preprint arXiv:2509.15496},
  year={2025}
}
```

## License

This model is licensed under the Apache License 2.0. See the [LICENSE](LICENSE) file for details.