Embrace the Unpredictability of Fractal Art

Enter the realm of unpredictability with Chaos Weave, a LoRA that brings chaotic beauty to any image. Chaos Weave unveils a different face with each use: sometimes enhancing images with intricate detail and swirling patterns, other times dissolving forms into mesmerizing fractal textures. At its most subtle, it leaves behind delicate traces; at its most intense, it transforms entire images into a kaleidoscope of fractal intricacies. See all the showcase images -- they were all generated with the same LoRA strength. I can't explain it.

I'll be honest; I'm convinced that training an AI on a gigantic fractal image broke it. This is the sixth version of the LoRA because I do not want to give in to the reality that there's something I don't understand going on there, so I keep tweaking it and trying it again. I've trained it from a few hundred steps to a few thousand with the same results. When I attempted to discuss it with my LLM friend (ChatGPT), he glitched and told me talking about it would violate his TOS.

Anyway, if you love the thrill of unpredictability, this will work out great. Whether adding subtle flourishes, intricate borders, or breaking down compositions into captivating fractal patterns, this LoRA takes each image through structured chaos.

Usage

To use the most recent version (and only published) version of the LoRA, use the following settings:

• Trigger word: difrctlchs; But honestly, the LoRA doesn't seem to care. I swear, you can use it and not use it on the same seed, and sometimes there is no difference, and other times the difference is extreme.

• Other tokens that work well: No tokens are needed to trigger the style. The model likes objects that are circles or that have swirls, waves, or curls. No particular words are needed; it just likes them when they are generated in the image.

• Lora Strength: A strength between 0.0 and 2.0 can be used and can have no effect or an extreme effect in equal measures.