How to create AI images locally in Windows, Mac or Linux for free in 2025

[Nautilus]

Using this guide you'll be able to create images using AI locally in windows, mac and linux for 100% free. So let's get started!

Step 1: Install python

Download & install python from here: https://www.python.org/downloads/release/python-3129/
NOTE: When installing in windows, click on 'Add to executable PATH' is in the installer

Step 2: Verify installation

Once the installation is finished, open a new terminal/command prompt and type python --version & press enter
It should show something like Python 3.12.9, this means that python was installed successfully  

(To open a command prompt in windows, press the windows key & search for cmd)

Step 3: Install CUDA Toolkit

Install CUDA Toolkit (for Nvidia graphic cards): https://developer.nvidia.com/cuda-downloads

Step 4: Install ComfyUI

ComfyUI is the base software we'll use to create AI images.
1. Go to https://github.com/comfyanonymous/ComfyUI and click on Download as ZIP
2. Extract the downloaded zip file to Desktop
3. Open a new command prompt in Desktop
4. Execute the following commands one by one:

 

Code:

cd ComfyUI

 

Code:

pip install -r requirements.txt

Step 5: Start ComfyUI

The installation is now complete. Execute the following command in cmd to start ComfyUI,:

Code:

python main.py

It'll take about a minute to start. Once it's started open http://127.0.0.1:8188 in your browser. You should see something like this:



Step 6: Download Image Generation model

Now to generate images, we need an AI model (checkpoint). There are a lot of checkpoints you can choose from, every checkpoint has something it specializes in like some checkpoints are optimized for creating realistic images, some are optimized for creating anime style images & so on.
A good starting point is this model: https://mega.nz/file/NjtiCbYB#5ck0RCsedo...2aJOjXnt6g

Once the model is downloaded, copy it to ComfyUI -> models -> checkpoints folder
Now go back to the browser window where comfyui is open & press R on your keyboard to refresh.

Step 7: Generate image

Click here & select 'epicrealismXL_vxviLastfameRealism.safetensors'



In this box, enter the image details like the things you want in the image (positive prompt)



In this box, enter the things that should not be in the image (negative prompt) like text, watermark



Here you can set the image size, we'll leave it at 512x512 for now.



We'll leave the other settings as default for now. Click on Queue to start image generation



It'll take a few minutes to generate the image depending on your GPU speed

Step 8: Result



All generated images are saved in the ComfyUI -> Output folder

That's all for this guide!

This is a basic guide to get you started with AI Image Generation & ComfyUI.
You can find more advanced guides & tutorials here: https://sdiffusers.com/Forum-Guides