Stable-Diffusion-React-App

A Simple React WebApp and a FastAPI backend that uses Stable Diffusion Pipelines to generate images.

Features

• Simple and intuitive UI
• Enter any prompt (maximum 77 tokens)
• Select model from various diffusion models
• All generated images are saved locally

To-Be-Implemented

• Save button to download generated image
• Generate images with more inference steps
• Enter custom heights and/or widths for images

How To Use

Add Token:

create 'auth_token.py' in /api and add variable "auth_token" with your huggingface token.

Run API : uvicorn api:app --reload

Run Client : npm start

API Routes

Generate

  GET /generate

Parameter	Type	Description
prompt	string	Required. The text prompt of an image

Select Model

  GET /selectmodel

Parameter	Type	Description
model	string	Required. The model to be used

References

• Stable Diffusion: https://huggingface.co/blog/stable_diffusion
• Diffusers: https://github.com/huggingface/diffusers/blob/main/README.md

Models

• Dream-like v1 Model: https://huggingface.co/dreamlike-art/dreamlike-diffusion-1.0
• Stable Diffusion v2 Model: https://huggingface.co/stabilityai/stable-diffusion-2
• Stable Diffusion v1.5 Model: https://huggingface.co/runwayml/stable-diffusion-v1-5
• Robo Diffusion: https://huggingface.co/nousr/robo-diffusion
• Mo Di Diffusion Model: https://huggingface.co/nitrosocke/mo-di-diffusion
• Waifu Diffusion Model: https://huggingface.co/hakurei/waifu-diffusion