Here's two different LORAs I trainedOne is via kohya, using <a target="_blank" rel="ugc" href="https://civarchive.com/models/23900?modelVersionId=95489">AnyLoRA - Checkpoint</a> as base model, which you can download in here.Another one was trained by using Pixai's model training feature, using Anything V5 as base model. (And you can use it online or download it in <a target="_blank" rel="ugc" href="https://pixai.art/model/1697066379180180314/1697071752532227312">here</a>)The pixai one is more flexible then the other, I'll try update kohya model once I have more times best weight are 0.7~0.8trigger words : vera_ongezellig, medium length hair, red scarf, coat, purple hairadditional trigger word: hair over one eye

Files with SHA256: 1b6a5e412c61698996d71fa1119e1295f9adaf896ba9d75e6678d53aa43af8f9