This LoRA was trained on 90 screenshots from all released episodes of “Lackadaisy” on the YouTube channel: https://www.youtube.com/@LackadaisyComic

Since the dataset contains only cats, this LoRA will predominantly generate cats. I did not bind the LoRA to the characters’ appearances using prompts. If you’re interested, let me know and I’ll create a new version of the LoRA with appearance binding all the way down to the individual characters (if it turns out to be possible within the style).

This model received technical support from ModelScope.