Model Information


Description
Personal practice works are intended for personal study and application. No payment is required and no commercial use is allowed.
Illustrious version:
Checkpoint: It works best with impasto style models, such as the "AnBan_Illus" series.
Lora weight:0.7
Vae: sdxl
Size: 1024*1536
Clip stop: -2
Steps: 25
Cfg: 5.5
Sampler: euler a
Upscale: 1.5, upscale modle: remacri_original, deniose: 0.3, also need Adetailer or face detailer to repair the face in high definition.
SD1.5 version:
Checkpoint: "Based64 v3",another one is a merge modle: "AbyssOrangeMix2 - NSFW" + "Async's MIX v6", the proportion is 0.5 each.
Lora weight:0.7
Vae: orangemixvaeReupload_v10, kl-f8-anime2
Size: 512*768 is best, but I might also be surprised when I enter a larger size
Clip stop: -2
Steps: 25
Cfg: 7
Sampler: euler a
Upscale: 1.5, upscale modle: remacri_original, deniose: 0.3
Almost Hyung-tae Kim Artstyle Lora
Model Details
- Type
- Model Addon
- Subtype
- LoRA
- Created
- Updated
- June 8, 2026