A personal mix. All preview images are t2i + hires.fix without any inpainting. It's kind of a mix of 2.5D and painting models, and have to use hires.fix for good results.
WARNING: It can do NSFW imgs.
This version come with my style lora in it. You will get a lot of similar face outputs. You can get the model with out my style at LibraMix.
All preview images are generated with below settings:
VAE: [kl-f8-anime2]
Clip Skip: 2
Sampling method: SDE Karras
Sampling steps: 30
Hires.fix: on
Upscaler: Remacri Upscaler
Hires steps: 10-12
Denoising strength: (0.4 - 0.5)
Upscale by: 2
WxH: 512x768
CFG Scale: 5 (5 - 7.5)
Check preview for negative prompt and embedding that I may use.
Description
Adjusting merge parameters on Camellia2.5D and dalcefo_painting!
FAQ
Comments (8)
Can u please bake the VAE ? pweeez
I changing vae base on the situation so idk which one to bake in tbh
what are u used Lora,vae?
great model
Any chance this would be updated for SDXL? This is my most favorite model. :)
The comments for this model mention "LibraMix", the same model without the style LoRA baked in. However, the provided link to this model just takes you to the same page as LibMix. Searching CivitAI for "LibraMix" comes up with no hits.
Has "LibraMix" been posted on CivitAI?