GitHub - Vchitect/RepVideo: The official implementation of ”RepVideo: Rethinking Cross-Layer Representation for Video Generation“ (original) (raw)

RepVideo: Rethinking Cross-Layer Representation for Video Generation

S-Lab, Nanyang Technological University1 Shanghai Artificial Intelligence Laboratory 2
†Equal contribution. ✉Corresponding Author.

Hits Hits Hits

🔥 Update and News

Installation

1. Create a conda environment and download models

conda create -n RepVid python==3.10 conda activate RepVid pip install -r requirements.txt

mkdir ckpt cd ckpt mkdir t5-v1_1-xxl wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/config.json wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/model-00001-of-00002.safetensors wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/model-00002-of-00002.safetensors wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/text_encoder/model.safetensors.index.json wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/added_tokens.json wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/special_tokens_map.json wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/spiece.model wget https://huggingface.co/THUDM/CogVideoX-2b/resolve/main/tokenizer/tokenizer_config.json

cd ../ mkdir vae wget https://cloud.tsinghua.edu.cn/f/fdba7608a49c463ba754/?dl=1 mv 'index.html?dl=1' vae.zip unzip vae.zip

2. Download Our latest Checkpoint

git-lfs clone https://huggingface.co/Vchitect/RepVideo/tree/main
# Then modify the "load" path in "sat/configs/inference.yaml" accordingly.

Inference

BibTeX

@article{si2025RepVideo,
  title={RepVideo: Rethinking Cross-Layer Representation for Video Generation},
  author={Si, Chenyang and Fan, Weichen and Lv, Zhengyao and Huang, Ziqi and Qiao, Yu and Liu, Ziwei},
  journal={arXiv 2501.08994},
  year={2025}
}

🔑 License

This code is licensed under Apache-2.0. The framework is fully open for academic research and also allows free commercial usage.

Disclaimer

We disclaim responsibility for user-generated content. The model was not trained to realistically represent people or events, so using it to generate such content is beyond the model's capabilities. It is prohibited for pornographic, violent and bloody content generation, and to generate content that is demeaning or harmful to people or their environment, culture, religion, etc. Users are solely liable for their actions. The project contributors are not legally affiliated with, nor accountable for users' behaviors. Use the generative model responsibly, adhering to ethical and legal standards.