A browser interface based on Gradio library for Stable Diffusion, input from .env files for an easy image creation.

• METHOD 1:

1. Install Python 3.10.6, checking "Add Python to PATH".
2. Install git.
3. On terminal:

git clone https://github.com/shiroppo/ai_image_env
cd ai_image_env
py -m venv venv
.\venv\Scripts\activate
pip install -r requirements.txt

• and then:

pip install torch==2.0.1+cu118 torchvision==0.15.2+cu118 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cu118

• METHOD 2:

1. Install Python 3.10.6, checking "Add Python to PATH".
2. Install git.
3. Download and run app_install.bat in the directory of your choice.

• Double click on app_gradio.bat on ai_image_env directory
• the interface is now open on the default browser, than select the .env/.yaml file that you prefer(in ./env/ directory)
• select the cycles(how many images you want)
• click "Submit"
• you can find the output in the ./image/today_date/ directory

• Download from hugginface or any similar website every checkpoint sd xl that you use and put them into ./models/Stable-diffusion/
• Download from hugginface or any similar website every checkpoint sd 1.5/2.0/2.1 that you use and put them into ./models/Stable-diffusion/
• Download from hugginface or any similar website every LoRA sd xl that you use and put them into ./models/Lora/
• Download from hugginface or any similar website every LoRA sd 1.5/2.0/2.1 that you use and put them into ./models/Lora/
• Download from hugginface or any similar website every VAE that you use and put them into ./models/VAE/
• Download from hugginface or any similar website every negative embeddings that you use and put them into ./embeddings/

In the prompt/negative prompt you can use the "prompt weighting":

- Example: "a grey cat++ playing with a ball----"

• in this example the cat(+) is MORE important that the ball(-)

• .env
• .yaml
• .png(created with ai_image_env)

Write __name__ in your prompt to get a random line from a file named name.txt in the ./wildcards directory.

In the selected .env:

• turbo_lcm=true
• for sd 1.5 checkpoint download the LoRA: https://huggingface.co/cleanmists/my/blob/61484b2d466b8a10349c5207981ea893e74f1c49/LCM_LoRA_Weights_SD15.safetensors and put it into ./models/Lora/
• for sd xl checkpoint download the LoRA: https://huggingface.co/shiroppo/sd_xl_turbo_lora/blob/main/sd_xl_turbo_lora_v1-128dim.safetensors and put it into ./models/Lora/

• v1.0: First Version
• v1.1: Added "Prompt" textbox, if empty the prompt is obtained from the selected *.env file
• v1.2: Added "Turbo Mode"

• ko-fi: (https://ko-fi.com/shiroppo)