ComfyUI-LVCDWrapper

★ 61

视频生成图像到视频LVCDComfyUI集成

为 LVCD 提供 ComfyUI 封装节点，便捷调用 LVCD + SVD 模型在 ComfyUI 中实现图像到视频的生成与管控，并自动处理模型路径与兼容性。

💡 在ComfyUI中用LVCD和SVD模型进行图像到视频生成

🍴 4 Forks💻 Python🔄 2024-09-30

🔗 GitHub 原文

📦

网盘下载

复制链接后前往夸克网盘下载

https://pan.quark.cn/s/b45acaa3411d

📦 requirements.txt

omegaconf>=2.3.0
clip
@
git+https://github.com/openai/CLIP.git
open-clip-torch>=2.20.0
pytorch-lightning>=2.0.1
timm>=0.9.2

📄 README

ComfyUI wrapper nodes for LVCD

https://github.com/user-attachments/assets/6d4d5afd-f8b1-499b-8872-d5d2f6929da4

Requires SVD model, seems to work best with the original one, but runs with 1.1 and XT as well, this is loaded normally from ComfyUI/models/checkpoints:

https://huggingface.co/stabilityai/stable-video-diffusion-img2vid

fp16 version:

https://huggingface.co/Kijai/LVCD-pruned/blob/main/svd-fp16.safetensors

LVCD model itself goes to ComfyUI/models/lvcd (autodownloaded if it doesn’t exist):

https://huggingface.co/Kijai/LVCD-pruned/blob/main/lvcd-fp16.safetensors

Original repo:

https://github.com/luckyhzt/LVCD