- Text/Video-to-Video 2B: https://huggingface.co/THUDM/CogVideoX-2b
- Text/Video-to-Video 5B: https://huggingface.co/THUDM/CogVideoX-5b
- Image-to-Video 5B: https://huggingface.co/THUDM/CogVideoX-5b-I2V
- Original Repository: https://github.com/THUDM/CogVideo
- Diffusers documentation: https://huggingface.co/docs/diffusers/en/api/pipelines/cogvideox
- Diffusers-TorchAO quantization benchmarks: https://github.com/sayakpaul/diffusers-torchao/
- Diffusers-Quanto example: https://gist.github.com/a-r-r-o-w/31be62828b00a9292821b85c1017effa
- HF CogVideoX Space: https://huggingface.co/spaces/THUDM/CogVideoX-5B-Space
Colab notebook examples:
- T2V: https://colab.research.google.com/drive/1pCe5s0bC_xuXbBlpvIH1z0kfdTLQPzCS?usp=sharing
- I2V: https://colab.research.google.com/drive/17CqYCqSwz39nZAX2YyonDxosVKUZGzcX?usp=sharing
- V2V: https://colab.research.google.com/drive/1comfGAUJnChl5NwPuO8Ox5_6WCy4kbNN?usp=sharing
- ComfyUI: https://github.com/kijai/ComfyUI-CogVideoXWrapper
- Pallaidium: https://github.com/tin2tin/Pallaidium
- CogVideoX-Fun: https://github.com/aigc-apps/CogVideoX-Fun
- CogStudio: https://github.com/pinokiofactory/cogstudio
- CogVideoX finetune for interior design: https://huggingface.co/bertjiazheng/KoolCogVideoX-5b
- Disney finetuning dataset: https://huggingface.co/datasets/Wild-Heart/Disney-VideoGeneration-Dataset
Other usage examples:
- Llama 3 + Flux + CogVideoX-I2V: https://gist.github.com/a-r-r-o-w/d070cce059ab4ceab3a9f289ff83c69c
- Naive video captioning for training with MiniCPM-V-2.6 + Llama 3: https://gist.github.com/a-r-r-o-w/4dee20250e82f4e44690a02351324a4a