LORA trained on 43 publicly available images of Alluux.
Trained on base SD 1.5, should work with most realistic models (works best with serenity v2).
Use ADetailer if details are lost and strength of 1 should work on most prompts.
Consider leaving a review if you like the model. All feedback is appreciated.
Tried training with just 23 face images, the results are good but it's less sensitive to changes from the original. Try adding more strength like ((blonde)) for it to work.
Description
FAQ
Comments (5)
it's very good it might need to have a couple of updates to make it perfect but otherwise amazing work
You should do Yeb next
Do you know any good sources for images?
Scoured all the usual spots, couldn't find a lot of high quality images.
Yungelderberry - v1.0 | Stable Diffusion LoRA | Civitai
Initial version with whatever images I could find
Will work on v2 if I find a better source
@manikb Awesome! Don’t know if any other spots but I’ll check it out.