VisualDiffusion v0.3.3 NEW MAKEGUDPICS UPDATE
A GUI for text2img diffusion, as a visual alternative to CLI and Jupyter Notebooks.
CLICK HERE FOR PROGROCK DIFFUSION or CLICK HERE FOR PROGROCK-STABLE
If you would like support me, the development of this GUI, or maybe you have a feature that you desperately want added, please consider supporting me and this software's development on my Patreon.
conda activate progrockdiffusion -or- conda activate prs
cd progrockdiffusion -or- cd prs
git clone https://github.com/KnoBuddy/visualdiffusion/
python visualdiffusion\vd.py
conda activate progrockdiffusion -or- conda activate prs
cd progrockdiffusion -or- cd prs
git clone https://github.com/KnoBuddy/visualdiffusion/
python3 visualdiffusion\vd.py
This will open either PRS or PRD depending on what is installed. If PRS is installed it will ask which GUI you would like to use, PRSGUI or MakeGUDPics.
cd progrockdiffusion\visualdiffusion -or- cd prs\visualdiffusion
git pull
Install PRS and ensure it is working and cd into directory. Follow the Instructions Here.
Adjust the sliders according to how much you want that portion of the prompt to effect the image. (Prompt weights are currently only working in the dev version of PRS.)
Follow the Instruction Here.
Open up a command line window (shell) and activate the conda environment like normal, and then run prdgui.py, from within the prs folder (NOT FROM WITHIN THE VISUAL DIFFUSION FOLDER). Please familiarize yourself with ProgRock-Stable's Settings.
You can add as many prompts as you please. The "Save Settings" button saves the settings to gui_settings.json. Make sure all the text boxes having an appropriate value. I also recommend naming your batch. If you need to cancel the render, you will have to kill the shell window you used to open the GUI, or CTRL-C to keyboard exception out.
Follow the Instruction Here.
Open up a command line window (shell) and activate the conda environment like normal, and then run prdgui.py, from within the progrockdiffusion folder (NOT FROM WITHIN THE VISUAL DIFFUSION FOLDER). Please read the settings file if you are unfamiliar with PRD.
You can add as many prompts as you please all with different weights. The "Save Settings" button saves the settings to gui_settings.json. Make sure you select at least one CLIP Model, as well as all the text boxes having an appropriate value, some can be left empty. I also recommend naming your batch. If you need to cancel the render, you will have to kill the shell window you used to open the GUI.
The final text box is for extra shell arguments. They will be parsed after the initial command:
python prd.py -s gui_settings.json <YOUR ARGS GO HERE>