By this point, you would literally have to live under a rock with no internet service to not know about AI image generators. In the past weeks, we’ve been flooded with incomprehensibly good and mind-numbingly bad AI-generated images. Midjourney and Dall E have been taking center stage during this digital circus and both are powerful in their own right.
Now we have Stable Diffusion (SD), an open source, free project that is so good even Midjourney has integrated it into their latest beta with amazing results. It’s been available for download at GitHub since its release but not all of us have the time or expertise to implement the thing from its GitHub form. All of that has changed, as we knew would happen, with standalone apps (GUIs), also free, now available to run Stable Diffusion from your local computer with no installation necessary.
Just download, unzip, and run the exe. This can be done from any location on the computer as it is not installed. To “uninstall” it… just delete it.
No monthly fees, no fast or slow time… just you, your computer, and your imagination… the only friends you’ve ever really had anyway. OK… maybe I should just speak for myself on that but it’s so cool to just run some images while waiting for a render to complete or a project to download.
I’m not only going to hook you up with one but two standalone GUIs (apps) that are tailored to the user that just wants to use. The only technical stuff is concentrating on the inputs, that magic you do when you make an AI-generated image. Those inputs can be a few words or a small essay. Whatever you wish to try because these two apps only cost computing time and space… on your own pc.
The first Stable Diffusion app I found was the sexily named: Stable Diffusion GRisk GUI 0.1. Hey, I’m sure it’s sexy if you are into code but as far as an outsider looking in the long name with the word RISK just didn’t sit right with me. I know… no self-respecting hacker would put RISK in the name of the program that had nefarious intentions. Or would they? After overthinking it for way too long I scanned it, opened it, and wondered why I was even worried about it.
It's a simple GUI with easy controls that implement the Stable Diffusion generator.
You’ll see two windows, one informational and the other the input window. You type, copy, or paste your input into the input text box. Set a few parameters and press the Render button to start the process.
There is no visual feedback with this GUI as the image is saved into the folder of your choice or the default location. You need to navigate to that location and open a folder to see the results. You can also generate multiple images by copying and pasting the same command to a new line. If you want 4 images rendered, then copy and paste the input four times in the input window before you press the render button. You will get 4 variations of the prompt this way.
Since it’s a local copy you can run as many renders as your heart desires or until you run out of storage space. It can be that addictive and now, without monthly cost, even more so.
Get the GRisk Stable Diffusion GUI here.
https://grisk.itch.io/stable-diffusion-gui
The second free SD GUI is the equally sexy named NMKD Stable Diffusion GUI. This variation is worth having because it uses img2img and allows the user to load a reference image into the interface with controls over how much influence it has. It also show you the end results in a right hand window.
Just like GRisk this is a self-contained file that only requires running the .exe file within the extracted folder. No installation is required. A word of warning though. Unlike the GRisk this one takes a while to set itself up once you run it the first time. It has some housekeeping chores and downloading to do, and it took long enough that I kept myself occupied with other tasks. You might think it is stuck but just wait it out.
Get the NMKD GUI for Stable Diffusion at the link below:
Anyway… here are two ready-made interfaces that you extract and click on the exe file to invoke. After that, it’s all fun and games. So much so you might be surprised at how many images you start generating since you can try just about anything that comes to mind.
Resource for Stable Diffusion user-generated images and keywords:
M.D. McCallum, aka WarLord, is an international award-winning commercial graphics artist, 3D animator, published author, project director, and webmaster with a freelance career that spans over 20 years. Now retired, M.D. is currently working part-time on writing and select character development projects. You can learn more about MD on his website.
