--- license: apache-2.0 --- # Video-BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generation
[📖 Paper](https://tacossp.github.io/BLADE-Homepage/) | [🚀 Homepage](https://www.google.com/search?q=%23-quick-start) | [💾 Models](https://huggingface.co/GYP666/VIDEO-BLADE) | [📖 中文阅读](README_zh.md)
Video-BLADE is a data-free framework for efficient video generation. By jointly training an adaptive sparse attention mechanism with a step distillation technique, it achieves a significant acceleration in video generation models. This project combines a block-sparse attention mechanism with step distillation, reducing the number of inference steps from 50 to just 8 while maintaining high-quality generation. ## 📢 News - **[Aug 2024]** 🎉 The code and pre-trained models for Video-BLADE have been released\! - **[Aug 2024]** 📝 Support for two mainstream video generation models, CogVideoX-5B and WanX-1.3B, is now available. - **[Aug 2024]** ⚡ Achieved high-quality video generation in just 8 steps, a significant speedup compared to the 50-step baseline. ## ✨ Key Features - 🚀 **Efficient Inference**: Reduces the number of inference steps from 50 to 8 while preserving generation quality. - 🎯 **Adaptive Sparse Attention**: Employs a block-sparse attention mechanism to significantly reduce computational complexity. - 📈 **Step Distillation**: Utilizes the Trajectory Distillation Method (TDM), enabling training without the need for video data. - 🎮 **Plug-and-Play**: Supports CogVideoX-5B and WanX-1.3B models without requiring modifications to their original architectures. ## 🛠️ Environment Setup ### System Requirements - Python \>= 3.11 (Recommended) - CUDA \>= 11.6 (Recommended) - GPU Memory \>= 24GB (for Inference) - GPU Memory \>= 80GB (for Training) ### Installation Steps 1. **Clone the repository** ```bash git clone https://github.com/Tacossp/VIDEO-BLADE cd VIDEO-BLADE ``` 2. **Install dependencies** ```bash # Install using uv (Recommended) uv pip install -r requirements.txt # Or use pip pip install -r requirements.txt ``` 3. **Compile the Block-Sparse-Attention library** ```bash git clone https://github.com/mit-han-lab/Block-Sparse-Attention.git cd Block-Sparse-Attention pip install packaging pip install ninja python setup.py install cd .. ``` ## 📥 Model Weights Download ### Base Model Weights Please download the following base model weights and place them in the specified directories: 1. **CogVideoX-5B Model** ```bash # Download from Hugging Face git lfs install git clone https://huggingface.co/zai-org/CogVideoX-5b cogvideox/CogVideoX-5b ``` 2. **WanX-1.3B Model** ```bash # Download from Hugging Face git clone https://huggingface.co/Wan-AI/Wan2.1-T2V-1.3B-Diffusers wanx/wan1.3b ``` ### Pre-trained Video-BLADE Weights We provide pre-trained weights for Video-BLADE: ```bash # Download pre-trained weights git clone https://huggingface.co/GYP666/VIDEO-BLADE pretrained_weights ``` ### Weight Directory Structure Ensure your directory structure for weights is as follows: ``` VIDEO-BLADE/ ├── cogvideox/ │ └── CogVideoX-5b/ # Base model weights for CogVideoX ├── wanx/ │ └── wan1.3b/ # Base model weights for WanX └── pretrained_weights/ # Pre-trained weights for Video-BLADE ├── BLADE_cogvideox_weight/ └── BLADE_wanx_weight/ ``` ## 🚀 Quick Start - Inference ### CogVideoX Inference ```bash cd cogvideox python train/inference.py \ --lora_path ../pretrained_weights/cogvideox_checkpoints/your_checkpoint \ --gpu 0 ``` **Argument Descriptions**: - `--lora_path`: Path to the LoRA weights file. - `--gpu`: The ID of the GPU device to use (Default: 0). **Output**: The generated videos will be saved in the `cogvideox/outputs/inference/` directory. ### WanX Inference ```bash cd wanx python train/inference.py \ --lora_path ../pretrained_weights/wanx_checkpoints/your_checkpoint \ --gpu 0 ``` **Output**: The generated videos will be saved in the `wanx/outputs/` directory. ## 📊 Project Structure ``` VIDEO-BLADE/ ├── README.md # Project documentation ├── requirements.txt # List of Python dependencies │ ├── cogvideox/ # Code related to CogVideoX │ ├── CogVideoX-5b/ # Directory for base model weights │ ├── train/ # Training scripts │ │ ├── inference.py # Inference script │ │ ├── train_cogvideo_tdm.py # Training script │ │ ├── train_tdm_1.sh # Script to launch training │ │ ├── modify_cogvideo.py # Model modification script │ │ └── config.yaml # Training configuration file │ ├── prompts/ # Preprocessed prompts and embeddings │ └── outputs/ # Output from training and inference │ ├── wanx/ # Code related to WanX │ ├── wan1.3b/ # Directory for base model weights │ ├── train/ # Training scripts │ │ ├── inference.py # Inference script │ │ ├── train_wanx_tdm.py # Training script │ │ ├── train_wanx_tdm.sh # Script to launch training │ │ └── modify_wan.py # Model modification script │ ├── prompts/ # Preprocessed prompts and embeddings │ └── outputs/ # Output from training and inference │ ├── utils/ # Utility scripts │ ├── process_prompts_cogvideox.py # Data preprocessing for CogVideoX │ ├── process_prompts_wanx.py # Data preprocessing for WanX │ └── all_dimension_aug_wanx.txt # Training prompts for WanX │ ├── Block-Sparse-Attention/ # Sparse attention library │ ├── setup.py # Compilation and installation script │ ├── block_sparse_attn/ # Core library code │ └── README.md # Library usage instructions │ └── ds_config.json # DeepSpeed configuration file ``` ## 🤝 Acknowledgements - [FlashAttention](https://github.com/Dao-AILab/flash-attention), [Block-Sparse-Attention](https://github.com/mit-han-lab/Block-Sparse-Attention): For the foundational work on sparse attention. - [CogVideoX](https://github.com/THUDM/CogVideo), [Wan2.1](https://github.com/Wan-Video/Wan2.1): For the supported models. - [TDM](https://www.google.com/search?q=https://github.com/Luo-Yihong/TDM): For the foundational work on distillation implementation. - [Diffusers](https://github.com/huggingface/diffusers): For the invaluable diffusion models library. ## 📄 Citation If you use Video-BLADE in your research, please cite our work: ```bibtex @article{video-blade-2024, title={Video-BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generation}, author={}, year={2024} } ``` ## 📧 Contact For any questions or suggestions, feel free to: - Contact Youping Gu at youpgu71@gmail.com. - Submit an issue on our [Github page](https://github.com/Tacossp/VIDEO-BLADE/issues).