Makes cheeky chibis. LyCORIS version of previously released LoRA.

How to use
Start with "chibi" and then add whatever else you want. For anything with realism mixed in I recommend putting "3D" and/or "realistic" in negatives and using DPM++ 2M Karras sampler since it leans more towards the drawing style.

Best used on NAI or other models with strong support for more simplistic drawn styles.
At very low strength I saw some signs indicating it could be used to improve general prompting and image quality in other models but it's not something I guarantee or you should expect this LoRA to do.

Misc
Dataset of 50 images trained at 512 resolution trained on NAI.
All images are untouched txt2img, no inpainting or helper embeddings used for quality transparence, e.g. - things should look even better if you do.
The training was done on a small subset of the artists images, as such the output of this LyCORS is not an accurate representation neither in quality or style of the source material. Go support the artist if you want the good stuff.