Add colorful fluff to your images or transform the world into vibrant fuzz. Check showcase images for suggested prompts, weights, etc.
The LoRA works best with people, animals and trees, but it can transform some objects and environments too.
Created using Konyconi's excellent guide:
https://civarchive.com/articles/771/tutorial-konyconi-style-lora-update
Description
FAQ
Comments (6)
Any tips for this? I've tried with 1.5, XL, and turbo XL models with weights between -10 and 10 using the keyword. The lora will change the image but nothing is fuzzy.
Oh i must not have tried a good 1.5 model, it does seem to work with most 1.5 checkpoints but no sdxl ones. I know the base model says 1.5 but the title says sdxl so I thought that would work too.
@Jamjamjam There's an SD1.5 version and a separate SDXL version, you can see which version you're using at the top left, below the lora name and tags
this needs a flux version! any hopes ? :-)
hmmm... sadly I don't have the dataset anymore but I can try
@martius72 thanks! This is one of my favorites. If possible, i greatly appreciate