Jagodibuja art style and two sub-styles, trained on a dataset of <s>1284</s> 1343 carefully curated images. Lowering the strength to 0.9-0.8 for the Pony version seems like a good idea.Main Trigger: JagoStyleSub Style Triggers: JagoStylePinup (soft shading, detailed) / JagoStyleFlat (cel-shaded, lower detail)Tagged Characters: JagoSophie, JagoErika, JagoArtur, JagoMegan, JagoElly, JagoClara, JagoJenny, JagoMila, JagoCarmen, JagoSolenne, JagoSonia, JagoJulia, JagoHirokoOptional Negatives: 3D, Watermark, Patreon logo, Web address, Text<pre><code>Update V2:
- JagoStyleComic is now JagoStyleFlat to avoid collisions with the concept of comics/comic panels. sub-styles are much better separated now
- Improved Dataset (up to Jan 2025), 1343 images, lots of tag fixes
- further improved nsfw
- Illustrious version added</code></pre>Works at any strength, any mode, profits a lot from ADetailer face, usually doesn't need highres fix. Only Lora to date that contains Clara and Jenny. Sub-style triggers can also be put into negatives to strengthen the filtering. Let me know how it works for you, especially if there's anything that needs improving!

JagoDibuja Style (PonyXL & Illust)

Trained on Illustrious BaseNot tested with NAI, recommend using other mixes like WaiNSFW, IllustriousToonMix or SteinIllustrious

Files with SHA256: b64737e099a6cc687b7fc7fed792ac3370e13d67d0c40a1178e13ad6364bdd98