In this blog post we are going to look at installing ComfyUI in just a few minutes so you can have Stable Diffusion up and running. You need to download the Checkpoint/Model file separately which is not included in the installer.


  • You have a Windows 10/11 PC
  • You have your own Nvidia GPU
  • You have download a Stable Diffusion model eg. SD 1.5 or see below some recommendations


  • Head over to the ComfyUI page in Github and only download the files from this location and no other.
  • Navigate down to the Installing > Windows section and click on the “Direct link to download“.
  • It will download a 7-zip file with extension .7z. You will need 7-zip to extract the contents.
  • Once extracted place the desired model (checkpoint file eg. .ckpt) in ComfyUI/models/checkpoints folder.
  • Now simply run the run_nvidia_gpu.bat (recommended) or run_cpu.bat

The command line will execute and generate a URL that you can now open in your browser.

Default graph of ComfyUI – simply click on Queue Prompt to generate images.

The default graph will load that is designed to run any Stable Diffusion model. You can select the model via the Load Checkpoint node, this should show all the model/checkpoints you download and placed in the checkpoint folder.

Here are a few models you should experiment with:

  • Protogen x5.3 – a very versatile model, jack of all trades pretty much
  • Deliberate – another great model – can get a bit naughty if you are not using negative prompts
  • Disney Pixar Cartoon A – great for CGI characters

The output images will be placed in the ComfyUI/outputs folder.

Video Tutorial


After a failed attempt at installing ComfyUI which I shared in my earlier post, I am very happy that I have ComfyUI working as a standalone installation. I will be expanding and building my graphs and sharing these on this blog so make sure you are following me and the blog on our X accounts.

If you'd like to support our site please consider buying us a Ko-fi, grab a product or subscribe. Need a faster GPU, get access to fastest GPUs for less than $1 per hour with