婕拉 驯龙女巫 Zyra

https://s2.xiaonaofu.top/material/1d3ca846-a2fd-4020-be8c-86081559f879.jpeghttps://s2.xiaonaofu.top/material/b6f90d32-9104-4787-994a-25a6001bfa56.jpeghttps://s2.xiaonaofu.top/material/81b09978-f659-425a-8005-70923d408bb4.jpeghttps://s2.xiaonaofu.top/material/5fb854ea-6973-4c54-a175-aacf93245717.jpeghttps://s2.xiaonaofu.top/material/04cc8583-0672-4019-90a6-59ba4f928581.jpeghttps://s2.xiaonaofu.top/material/f2959747-26ca-4c36-a7d9-abea7f882404.jpegdragon sorceress zyra

LoRA trained on Dragon Sorceress Zyra from League of Legends

requested by @PettankoPaizuri

models used

anime: YabaL_Mix V3, MeinaAlter

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

Dragon Sorceress Zyra: dragon sorceress zyra, 1girl

optional tags: dragon girl, dragon horns, tail, 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

Hide

发表回复

后才能评论