|
|
--- |
|
|
tags: |
|
|
- image-to-video |
|
|
- video-generation |
|
|
- personalized-video |
|
|
- identity-preservation |
|
|
pipeline_tag: image-to-video |
|
|
license: apache-2.0 |
|
|
base_model: |
|
|
- Wan-AI/Wan2.1-T2V-14B |
|
|
--- |
|
|
|
|
|
# Lynx: High-Fidelity Personalized Video Generation (GGUF by Vantage with AI) |
|
|
|
|
|
**Watch us at Youtube:** [@VantageWithAI](https://www.youtube.com/@vantagewithai) |
|
|
|
|
|
|
|
|
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 GGUF Quantized versions of the https://huggingface.co/Kijai/WanVideo_comfy/tree/main/Lynx: |
|
|
|
|
|
- **Lynx Ref. Full Model** (`full_ref`): Complete version with all advanced features and best performance |
|
|
- **Lynx IP Lite Model** (`lite_ip`): GGUF Quantized version of Lightweight IP model with fewer parameters, 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. |
