Ah, ComfyUI SDXL model merging for AI-generated art! That's exciting! Merging different Stable Diffusion models opens up a vast playground for creative exploration.
Here is my way of merging BASE models and applying LORAs to them in non-conflicting way using the ComfyUI (grab the workflow itself in the attachment to this article):
From the left to the right: there are four base models (green), which are merged pair-by-pair into one and ... in the same fashion (after they are merged in the very middle of the workflow), two LORAs independently (not one after another!) are being applied. As a intermediate result we now again have two almost identical models, but with a difference of the applied LORA. And finally, merging them gives us the output model, ready to use for our creative purposes.
As an example of this workflow you can see my model - RedOlives: https://civitai.com/models/283810