Stable Diffusion (SD) is one of the best ways for AI to make art. Since it is based on an open-source model, it has grown quickly and gained a lot of attention.
Stable Diffusion is free to use, and you can run it on your own computer or in the cloud.
To run Stable Diffusion on your own computer, you’ll need the following:
- At least 4 GB VRAM NVIDIA GPU (6 GB VRAM will be faster)
- 10 GB of local storage space
- A Linux, Windows 11, 10, 8.1, 8, or Mac operating system
But at the moment, most computers sold to consumers do not have GPUs. If this is the case, you can use Google Colab to run Stable Diffusion for free. Google Colab is just a place to write code in the cloud.
You can also use Stable Diffusion on DreamStudio, which is free. DreamStudio is a powerful creative tool that you can use in your browser. It is made by the same people who made Stable Diffusion.
DreamStudio is for people who want to run Stable Diffusion but don’t want to manually install Python modules, SD WebUI, SD checkpoints, and GFPGAN v1.4 checkpoints. But you can only get 200 free credits if you use your Discord account to sign in. When your free credits run out, you have to buy a membership to keep making AI images.
If you want to run Stable Diffusion on your own computer, you need to meet the above requirements. On top of that, every time you want to run the SD model, you have to go to the command-line tool or local web server.
Windows, Mac, and Linux Users Can Use UnstableFusion
UnstableFusion is a desktop front-end for Stable Diffusion (SD) that has important features like inpainting, img2img, and more.
UnstableFusion is a popular front-end app for running SD on Windows, Mac, and Linux. This tool makes it easy to run SD on the local machine by getting rid of the need to go to the command-line tool or local web server often and giving you an easy-to-use user interface (UI). But you still have to install Python, the Stable Diffusion model, and a few other parts.
Also Read: MacGruber 2: Important Details
How to Use UnstableFusion in Your Own Area?
- Use the pip command to install the dependencies that are needed. These dependencies are PyQt5, NumPy, PyTorch, Pillow, OpenCV-python, requests, flask, diffusers, transformers, and protobuf. Note: Use requirements-local GPU-win64.txt if you want to run StableDiffusion locally on Windows. Command: pip install -r requirements-local GPU-win64.txt
- Make an account on Hugging Face and an access token. Then, ask to see the StableDiffusion model.
- Copy this repository and run python unstablefusion.py.
Using UnstableFusion
- You will be given a square box where you can make images and change them. The square box will be on your right, and the options will be on your left.
- Enter the prompt in the “destination box” just below “Stable Diffusion Parameters.” Then, click the button that says “Generate.”
- You can use your keyboard to undo and redo the process. Press Ctrl+Z to undo, and Ctrl+Shift+Z to redo.
- You can add more detail to an image by using the “Increase/Decrease” buttons.
- Use “Scratchpad” to bring in images from elsewhere.
Diffusion Bee
Diffusion Bee is another front-end for Mac M1/M2 computers that make it easy to run Stable Diffusion.
Diffusion Bee is different from UnstableFusion in that you don’t have to install Python, the Stable Diffusion model, and other parts. It’s more like plug-and-play. Install the Diffusion Bee app, answer the question, and click “Generate.”
Since the application is not yet on the MAC App Store, you can download it from the Diffusion Bee website.
Also Read: Why All the Reviews for Windows Manufactured by Pella are Positive?
Conclusion
Stable Diffusion (SD) is one of the best ways for AI to make art. It is free to use, and you can run it on your own computer or in the cloud. UnstableFusion is a popular app for running SD on Windows, Mac, and Linux. If you’re running UnstableFusion on a Mac, you might want to install the dependencies that are needed to run it. Diffusion Bee is another front-end for Mac M1/M2 computers that make it easy to run Stable Diffusion on your Mac.
