[LoRA] Lidia Sobieska | Tekken

模型简介：
Lora trained on Lidia Sobieska from Tekken
tags: lidia sobieska, 1girl,
optional: scar across eye, tank top, black belt, white hair, white pants, blue eyes, red tank top
models used:
anime: CoffeeMix, PrimeMix
2,5D: MeinaUnreal - V3 (mostly used for the pictures), majicMIX fantasy - v2.0, PsyGon Animerge - v1.1, CamelliaMIx_2.5D_V2
realistic: fantasticmix_real, henmix_real
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, kl-f8-anime2, blessed2.vae (for anime models)
▣ Sampler: DPM++ SDE Karras or DPM++ 2M Karras
▣ Steps: 20
▣ Clipskip: 2
▣ Upscaler : img2img in comfyui using 4x-UltraMix_Smooth
▣ CFG Scale : 5 to 9 (mostly 7 and 8)
▣ LORA weight for txt2img: 0.6-1 (0.8 used in the pictures)
▣ 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

~~~~~~~~~~~~~~~~~~~~~~~o(*￣︶￣*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数：
Prompt: lidia sobieska, 1girl, scar across eye, tank top, black belt, white hair, white pants, blue eyes, fighting stance, kicking, solo, looking at viewer, potrait, indoor, tatami, sliding doors, shouji, sunbeam, (masterpiece:1.2, best quality)

Negative prompt: embedding:EasyNegative, embedding:bad_prompt_version2, embedding:ng_deepnegative_v1_75t, drawn by embedding:bad-artist, embedding:bad-hands-5, (worst quality, blurry, low quality:1.2), facial hair, nsfw, bad anatomy, text, 3 hands, multiple arms, watermark, stippling, craigslist, nipples, nipple throuch clothes, multiple girls, simple background, white background, noise, blurry background, blurry, signature, black background, cat ears, furry

Steps: 20, CFG scale: 7, Sampler: DPM++ SDE Karras, Seed: 883369565464149