莫甘娜 魔女 Morgana

https://s2.xiaonaofu.top/material/d071c132-93dd-4124-bb5d-b2788849ddb3.jpeghttps://s2.xiaonaofu.top/material/25c1b599-37e4-4a7f-9b63-1914d3b38629.jpeghttps://s2.xiaonaofu.top/material/1d2d1946-19d0-48ea-b15f-0af3fe9fda84.jpeghttps://s2.xiaonaofu.top/material/b1cab238-1ed5-44c4-a8dd-52b4ba7318ed.jpeghttps://s2.xiaonaofu.top/material/9bfb2c57-13b6-4d36-b68e-1ee672582a7a.jpeghttps://s2.xiaonaofu.top/material/e9201d42-5419-4f58-aa5c-6fc67e2eb83f.jpegcoven morgana

LoRA trained on Coven Morgana from League of Legends

models used

anime: MeinaAlter

2,5D: CamelliaMIx_2.5D_V2, MeinaUnreal – V3, HELLmix

realistic: fantasticmix_real

Coven Morgana: coven morgana, 1girl, dress, wings,

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

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

Steps: 20 to 35

Clipskip: 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

发表回复

后才能评论