Click to see where Colab generated images will be saved
Now you can set any count of images and Colab will generate as many as you set
Full tutorial for python and git installation with venv
Download and install Python 3.10.x (3.10.6, 3.10.9, 3.10.11) and git
- 3.10.9 > https://www.python.org/ftp/python/3.10.9/python-3.10.9-amd64.exe
- git > https://git-scm.com/downloads
- This installation won't affect or your break your other installations such as Automatic1111 Web UI
- Auto Installer Script : https://www.patreon.com/posts/85678961
- Instructions of auto installer shared in the post
- This installation won't affect or your break your other installations such as Automatic1111 Web UI
git clone https://github.com/FurkanGozukara/stable-diffusion-xl-demo
-
Register Hugging Face and login if you don't have an account already
-
Open below links and accept terms and conditions - now auto approve official weights - just type anything to the form fill sections
-
https://huggingface.co/stabilityai/stable-diffusion-xl-base-0.9
-
https://huggingface.co/stabilityai/stable-diffusion-xl-refiner-0.9
-
Generate your Hugging Face access token and save it somewhere
-
Enter inside cloned stable-diffusion-xl-demo folder
-
Follow the below commands 1 by 1. Watch video to see how I am making
python -m venv venv
cd venv
cd scripts
activate
cd..
cd..
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
pip install -U --pre xformers
pip install -r requirements2.txt
-
First edit app2.py and add your access_token. E.g. like below
-
access_token = "hf_ovfazhEBOXmTpUsfdfdshfdydZIfMNJVKrcZ"
-
The access token is only 1 time
-
After that you have to save below commands depending on your VRAM as .bat file and use them to start
-
Watch tutorial video again if you can't make it. All steps are shown
Low VRAM (12 GB and Below)
@echo off
set VENV_PATH=stable-diffusion-xl-demo\venv\Scripts\activate.bat
call "%VENV_PATH%"
set PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:128
set ENABLE_REFINER=false
python stable-diffusion-xl-demo\app2.py
pause
High VRAM (24 GB and above)
@echo off
set VENV_PATH=stable-diffusion-xl-demo\venv\Scripts\activate.bat
call "%VENV_PATH%"
set PYTORCH_CUDA_ALLOC_CONF=max_split_size_mb:256
set ENABLE_REFINER=true
python stable-diffusion-xl-demo\app2.py
pause