Skip to content

Reverse engineered API of Stable Diffusion XL 1.0 ( Midjourney Alternative ), A text-to-image generative AI model that creates beautiful 1024x1024 images.

License

Notifications You must be signed in to change notification settings

KoushikNavuluri/stable-diffusion-xl-api

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Stable Diffusion XL ( API )

Reverse engineered API of Stable Diffusion XL 1.0 ( Midjourney Alternative ) via https://replicate.com/ , A text-to-image generative AI model that creates beautiful 1024x1024 images.

Table of Contents

Prerequisites

To use this API, you need to have the following:

Python installed on your system requests library installed

  pip install requests

Installation

To use the Claude AI Unofficial API, you can either clone the GitHub repository or directly download the Python file.

Terminal :

pip install sdxl

or

Clone the repository:

git clone https://github.com/KoushikNavuluri/stable-diffusion-xl-api.git

Usage

Import the claude_api module in your Python script:

from sdxl import ImageGenerator
  • Next, you need to create an instance of the ImageGenerator class:
client = ImageGenerator()

Send Prompt to generate image

images = client.gen_image(
    "Vibrant, Headshot of a serene, meditating individual surrounded by soft, ambient lighting.")
print(images)

Output

Example Images Generated

Advanced Generation using parameters

#Parameters set to their default values
images = client.gen_image(prompt=
    "Vibrant, Headshot of a serene, meditating individual surrounded by soft, ambient lighting.",count=1, width=1024, height=1024, refine="expert_ensemble_refiner", scheduler="DDIM", guidance_scale=7.5, high_noise_frac=0.8, prompt_strength=0.8, num_inference_steps=50)
print(images)

List of parameters

  *   prompt = Input text prompt
  *   width  = Width of output image(max:1024)
  *   height = height of output image(max:1024)
  *   count  = Number of images to output. (minimum: 1; maximum: 4) 
  *   refine = Which refine style to use ( no_refiner or expert_ensemble_refiner or base_image_refiner )
  *   scheduler = scheduler (valid_schedulers = ["DDIM" or "DPMSolverMultistep" or "HeunDiscrete" or "KarrasDPM" or "K_EULER_ANCESTRAL" or "K_EULER" or "PNDM"])
  *   guidance_scale = Scale for classifier-free guidance (minimum: 1; maximum: 50) 
  *   prompt_strength = Prompt strength in image (maximum: 1) 
  *   num_inference_steps = Number of denoising steps (minimum: 1; maximum: 500) 
  *   high_noise_frac = for expert_ensemble_refiner, the fraction of noise to use (maximum: 1)

CLI Version

For cli version you can check example folder in this repository (filename:cli.py)

How to:

python main.py "beautiful landscape with two kittens,realistic,4k" --count 1 --width 1024 --height 1024 --refine expert_ensemble_refiner --scheduler DDIM --guidance_scale 7.5 --high_noise_frac 0.6 --prompt_strength 0.9 --num_inference_steps 40

Disclaimer

This project provides an unofficial API for Replicate's Stable Diffusion XL and is not affiliated with or endorsed by Replicate or Stable Diffusion. Use it at your own risk.

License

This project is licensed under the MIT License - see the LICENSE file for details.

About

Reverse engineered API of Stable Diffusion XL 1.0 ( Midjourney Alternative ), A text-to-image generative AI model that creates beautiful 1024x1024 images.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages