I changed my training parameters for this one so it might perform slightly differently than before. However, I manage to reduce the size to 36MB for file size.
If you want her to smile, just add "smile" to the prompt.
If you want a smile with teeth, just add "smile, teeth" to the prompt.
For how I train LORAs, see description of JAV Anzai_RI○N | Stable Diffusion LORA | Civitai
Must turn on Hires.fix
Hires steps:17, denoising:0.5+
Use DPM++ 2M Karras or DPM++ SDE Karras
Strength: 0.6-0.7
She might look like someone from real life, but she is not what you think. She is a totally fictional character.
Checkpoint used for samples: Basil Mix and KanPiroMix
VAE: vae-ft-mse-840000 from StabilityAI
Legal & Risk
It is prohibited to use this model for commercial purposes and any scenarios of illegal acts and purposes.
本人原则上不支持使用此模型生成任何成人内容。用户在使用本模型时,请务必遵守当地法律法规,勿侵犯他人的名誉权、隐私权或肖像权。
Description
Large version as all my other LoRAs.
FAQ
Comments (9)
For some reason I'm not getting notifications from your posts anymore, kinda annoying, good thing I checked your profile! Can't wait to try it out!
omg! sooo good
thanks so much
Amazing work, seeing all these beautiful women get added. I hope you get to Kaho Shibuya as well.
Thanks for all your effort and tutorial. May I know about your final loss after training? Does it truly relevant to the real quality?
谢谢大佬~
Just saw there is a large version now, what is exactly the difference?
Ever thought off... doing Sola Aoi?
So good.Can you share this to https://tensor.art/
Thanks alot. could help many guys