Powered by LCM - https://latent-consistency-models.github.io/
python3 -m venv pwd
/venv
source venv/bin/activate
pip3 install -r requirements.txt
python3 rtsd.py
As you move the sliders the image will update immediately. Put in two prompts like "cat" and "dog", or "Emma Watson" and "Tom Cruise". As you slide Merge Ratio you'll blend these prompts. Guidance and steps are typical to SD. More steps improve the image.
The seed if fixed to a initial value when you start the app. This produces better consistancy to compare images as you adjust the sliders. However, you can click "New seed" to change the seed.