body shape, Skin style Lora
I think you can get better results by using your current LoRA together with mamako-milf - v1.0 | Anima LoRA | Civitai
recommended strength :
v1.0~4.0 : 0.7 or lower
v4.1 : 0.4~1.0recommended trigger word : s-line, shiny skin(optional)
v4.1
This version delivers more stable and highly detailed images compared to previous versions.
Trained on Anima Base Model.
Style is exactly same as v4.0, fixed some issues
Quality improvement
Changed training resolution from 1280 to 1536
Improved quality for hand and foot
Fix issue where mask overlay appeared in generated images
v4.0
Trained on Anima Official release
v3.0
Trained on anima preview3 and Dataset partially changed
recommended strength : 0.4~0.6
an issue may occur where faces are erased or covered by white circles (above 0.7)
v2.0
Quality improved.
No more headless generations
v1.0
Headless image generated when strength abvoe 0.75
Description
Improved overall quality
FAQ
Comments (2)
This style is good, but it has strong bias for full body pictures. I'm usually not against that, in fact mostly I prefer it. Maybe I should be more detailed on the picture composition to compensate for this.
Not entirely sure, this may be due to the dataset being trained mostly on full-body images. I’ll keep that in mind for the next version. Thank you for your feedback!