Custom ComfyUI Workflow with the Krita AI Plugin #

Installation #

First visit the project’s releases page on GitHub, then download the latest zip file and make sure you click the first zip for it instead any of the Source code archives!

An image showcasing how to download krita-ai-diffusion.

??? #

Click on the workspace selection button on the top left side of the Docker.

An image showcasing how to connect to a custom ComfyUI workflow with krita-ai-diffusion.

Essential Nodes For Working with Krita Plugins in ComfyUI #

You will need comfyui-tooling-nodes installed in your custom_nodes folder.

cd ~/ComfyUI/custom_nodes
git clone https://github.com/Acly/comfyui-tooling-nodes

If you are going to use the Translate Text node you will also need to install argostranslate

pip install argostranslate

There are two of these nodes, the Krita Output node which you just need to connect to the VAE Decode’s image output.

A screenshot of ComfyUI demonstrating the Krita Output node.

Plus the Krita Canvas node, with it’s width and height output connected to anything that requires the total width and height of the images you are trying to generate, like the Empty Latent Image node or the CLIPTextEncodeSDXL and the seed input should be hooked to KSampler’s seed.

A screenshot of ComfyUI with the Krita Canvas node connected to the Empty Latent Image and other nodes.

A quick reminder that in order to convert a widget to an input all you need to do, is right click on the node and find the widget under Convert Widget to Input like this:

If you are using the new frontend you can also just drag the seed noodle directly over the widget. 🐺