Generating by prompts using SDXL-turbo

Introduction

We implement a module that generates images based on user-defined (text) prompts. We use the pretrained Stable Diffusion XL-Turbo provided by stabilityai at HuggingFace, while preparing the module.

Setting Up the Environment

Using Conda (recommended)

Install Conda, if not already installed.

Clone the repository:

git clone https://github.com/byrkbrk/generating-by-prompt-sdxl-turbo.git

Change the directory:
```
cd generating-by-prompt-sdxl-turbo
```

Create the environment:

conda env create -f generating-by-prompt-sdxl-turbo.yaml

Activate the environment:

conda activate generating-by-prompt-sdxl-turbo

Using Pip

Download & install Python (version==3.11)

Clone the repository:

git clone https://github.com/byrkbrk/generating-by-prompt-sdxl-turbo.git

Change the directory:
```
cd generating-by-prompt-sdxl-turbo
```
Install packages using pip:
```
pip install -r requirements.txt
```

Generating Images

Check it out how to use:

python3 generate.py -h

Output:

Generate images by prompts using SDXL-turbo

positional arguments:
  prompt                Text prompts that be used for generating

options:
  -h, --help            show this help message and exit
  --num_inference_steps NUM_INFERENCE_STEPS
                        Number of inferences steps that be used during
                        inference. Default: 5
  --device DEVICE       Device name used for inferece. Default: None
  --height HEIGHT       Height of the generated image. Default: 512
  --width WIDTH         Width of the generated image. Default: 512

Example usage

python3 generate.py\
 "an image of turtle in Picasso style"\
 "an image of turtle in Camille Pissarro style"\
 --num_inference_steps 10

The output images seen below (left: Picasso style, right: Pissarro style) will be saved into generated-images folder.

Generating Images by Gradio

To run the gradio app on your local computer, execute:

python3 app.py

Then, visit the url http://127.0.0.1:7860 to open the interface.

Example usage

See the display below for an example usage of the module via Gradio for the prompt 'a picture of a lion in Claude Monet style' with inference steps of 5.