DEPRECATED

This is repository is no longer maintained, please consider using A1111 fork instead.

diffusion-ui-backend

Gradio backend for the diffusion-ui web frontend using an unified Stable Diffusion diffusers pipeline

The gradio interface provides an API to generate images with Stable Diffusion for:

• text-to-image
• image-to-image
• inpainting

Documentation

The documentation is available here

Installation

Detailled installation instructions are available in the documentation.

First install pytorch with cuda support (if you have a NVIDIA GPU):

conda install pytorch torchvision torchaudio cudatoolkit=11.6 -c pytorch -c conda-forge

Then install diffusionui and its dependencies:

pip install diffusionui

First Usage

The first time, you have to download the model:

• create an account on https://huggingface.co
• Click on this page to accept the LICENSE
• generate a token in your settings
• login on your console with huggingface-cli login
• then download the model with:

# using the low-memory model (for GPUs with low VRAM)
diffusionui --low-mem --download-model

# or using the full model
diffusionui --download-model

Usage

Once the model has been downloaded, you can start the backend by running:

# For the low-memory model
diffusionui --low-mem

# For the full model
diffusionui

It should produce an local URL for the gradio interface:

Running on local URL:  http://127.0.0.1:7860/