Stable Diffusion Web UI

The following post is a collection of notes and links on SDUI usage and modification.

Adding stabilityai/stable-video-diffusion-img2vid-xt to Automatic1111’s SDUI. See:

  1. https://github.com/continue-revolution/sd-webui-animatediff/blob/forge/master/docs/how-to-use.md#you-have-a1111-and-you-know-git — Adding forge to sdui
  2. https://www.reddit.com/r/StableDiffusion/comments/1amvccg/stable_video_diffusion_in_automatic1111/ — Reddit suggestion to use forge to implement svdxt
  3. https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt — Stable Video Diffusion XT
  4. https://www.toolify.ai/ai-news/unleash-the-power-of-stable-diffusion-webui-forge-with-automatic-1111-1486458 — Info on Forge

After installing the appropriate virtual environment:

python3 -m venv sdui_env
pip install -r requirements.txt

and the required modules to allow forge to function, you need to git pull the svd model into the ./models/svd directory:

cd models
cd svd
git pull https://huggingface.co/stabilityai/stable-video-diffusion-img2vid-xt

Adding zero123 to forge

  1. https://huggingface.co/stabilityai/stable-zero123 — repo used

So the SVD and z123 implementations works, now for testing the 2 open methods for constructing 3d models. (z123 only generates a few stills at different angles)

Leave a Reply

Your email address will not be published. Required fields are marked *