This LORA model is trained with Hassaku (SD1.5) and Yamer's Anime (SDXL). it works pretty well with any anime/illustration models.
SD1.5
Recommend:
Image Size: 512x512 or 512x768
Steps: 20-30
Sampler: DPM++ 2M Karras
CFG scale: 7-11
Weight: 0.4-0.6
SDXL
Recommend:
Image Size: 1024x1024 or 1024x1536 or 1080x1350
Steps: 20-30
Sampler: DPM++ 2M Karras
CFG scale: 7-11
Weight: 0.4-0.6
Trigger Word: manhwa-artstyle
** I use 4x-Ultrasharp to upscale the images
Description
Support SD1.5