While training on SDXL (or PDXL) takes more time, it seems to have higher chance (compared to sd1.5) to succeed at making something useful with very limited dataset. So I tried with a light novel waifu who has only a few official illustrations.

DarkMel (Black Summoner)

Trained on <s>LN illustrations and some generated nudes</s> images generated from previous iterations and a few of LN illustrations.

faux-darkmel-mixed12-illu-came-rank128a64-lora.safetensors

illu-mixed

v1.0 pony

v0.1 pdxl

creator: fauxWhile training on SDXL (or PDXL) takes more time, it seems to have higher chance (compared to sd1.5) to succeed at making something useful with very limited dataset. So I tried with a light novel waifu who has only a few official illustrations.

Trained on images generated from previous iterations and a few of LN illustrations.


Files with SHA256: ecae6b5c9c141f3b0eec5309d291edcec28ecb3eb5dddc010a92ab99ad2e2f5f