Second LoRA model with only a few datasets, at the moment, this is what I used:
Model: Chilloutmix_NiPrunedFp32Fix (best result)/ Basil_mix_fixed
Sampling: DPM++ SDE Karras / DPM++ 2M Karras
Sampling: 30
Hires steps: mostly not being used (feel free to try)
Size: 512x768
Upscaler: Latent(Bicubic antialiased)
Denoising: 0.5-0.6
CFG Scale: 6.5
LoRA strength: 0.6-0.89
Face restoration: use GFPGAN 0.5
Use the usual VAE
Hires fix is optional
Feel free to experiment with other models/settings and leave a review.
She might look like someone from real life, but she is not what you think. She is a totally fictional character.
If you feel like supporting, do leave a tip at my kofi
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。
I generally do not support the use of this model to generate any adult content. When using this model, users should abide by local laws and regulations and shall not violate any other's right to reputation, privacy or portrait.
私は一般的に、このモデルを使用して成人コンテンツを生成することをサポートしません。 このモデルを使用する場合は、ユーザーは当地の法律や規制を守り、他人の名誉権、プライバシー権または肖像権を侵害しないようにしなければなりません。
Description
Refer to sample images prompts, and you should produce consistent results based on that.
Feel free to test out with different settings and let me know if you figured out better results :)
No plan to add v2 at the moment, unless necessary.
LoRA trained with smiling face mostly, so it handled smile expression better.
No trigger words needed.