LoHA trained on Iki hiyori, Yato, Yukine from Noragami

gif idea inspired by @Elluna Rem and Ram

models:

anime: MeinaAlter

2,5D: MeinaUnreal

tags:

Iki hiyori 1: iki hiyori, 1girl, pink scarf, long hair, brown hair, coat

Iki hiyori 2: iki hiyori, 1girl, long hair, school uniform, serafuku

u could also add a tail but its not that consistent

Yato 1: yato, 1boy, track suit, bandana

to get the bandana consistently increase the weight of the LoHA to 1 or just be lucky :D

Yato 2: yato, 1boy, male focus, japanese clothes, short ponytail

Yukine 1: yukine, 1boy, fur-trimmed jacket, blonde hair

Yukine 2: yukine, 1boy, hoodie, beanie, blonde hair

negative embeddings: embedding:bad-hands-5, embedding:ng_deepnegative_v1_75t, embedding:bad-artist, embedding:EasyNegative

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• VAE: orangemix.vae (for 2,5D pictures), vae-ft-mse-840000-ema-pruned (for realistic models), kl-f8-anime2, blessed2.vae (for anime models)

• Sampler: DPM++ SDE Karras or DPM++ 2M Karras

• Steps: 20 to 35

• Clipskip: 1 or 2

• Upscaler : img2img in comfyui using 4x-UltraMix_Smooth

• CFG Scale : 5 to 9 (mostly 7 and 8)

• LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.8)

• Denoise strength for img2img: 0.4 to 0.7

If you want the exact same settings, just drag and drop the picture into ComfyUI

for the pictures i used comfyui