Stable Diffusion API with SimianLuo/LCM_Dreamshaper_v7

Overview

This project provides an API endpoint for generating images using Stable Diffusion with the SimianLuo/LCM_Dreamshaper_v7 model. It supports image-to-image generation with customizable parameters.

API Endpoint

POST /generate Generates images based on the given parameters

Parameters:

• reset (bool): Set to True to generate an image from the original. Default is False.
• str (float): Strength of the image transformation. Controls how much the original image is altered. Default is 0.1.
• gui (float): Guidance scale to influence the prompt’s effect. Default is 10.0.
• num (int): Number of inference steps to run. Default is 5.
• prompt (string): The text prompt to guide the image generation. Default is a predefined prompt.
• count (int): Number of images to generate. Default is 1.
• width (int): Width of the generated image in pixels. Default is 128.
• height (int): Height of the generated image in pixels. Default is 128.

Example Request:

GET /generate?prompt=Devil-man&str=0.5&gui=7.5&num=10&count=16&width=512&height=512

This will generate 16 devil man images of size 512x512, each with a strength of 0.5, guidance scale of 7.5, and 10 inference steps from the previous.

Installation

1. Clone the repository:

   git clone https://github.com/JohnnyDeng6/sd3-server.git
   cd sd3-server

2. Install dependencies:

   pip install -r requirements.txt

3. Run the server:

python3 main.py

The API will be available at http://localhost:8888/generate