UPDATE (26/10/2026):

MOSS test version trained using default settings on Civitai trainer as wanted to see how lora training was on the new model, so might be a bit amateurish. Trained using the same dataset and captions as Anima / Illustrious versions but seems to work fine with natural language prompts. Have a play around with the settings to get illustrations or realism.

UPDATE (17/05/2026):

Retrained on Anima Base version. Training material stayed the same.

Update 13/04/2026: Anima version created based on Preview 3. Still experimenting with creating LORAs with this model.

Illustrious model for Conqueror Lilias from Epic Seven. This is my 1st attempt at a LORA and used this character as no IL Lora exists for this variant of her.

Trained on Mac on the Draw Things app using the Illustrious-XL model for 1750 steps. I used 27 images with the vast majority of them being 768x768 pix images. Training took around 1.5-2 hrs as I picked modest options just as a test.

Based on my tests, it seems to work fine for a lot of the Illustrious models, but maybe not so much for some of the merges (especially if they are merging with Noob or another model family). Would advise using face adetailer as the face doesn't always come out perfect. Tried it with some art-style loras and had some success when the strength of this lora and the art-style lora was put to 0.95 (based on two art style LORAs I checked).