公告：新站新QQ群：281257170

1、本站为非营利性小站，其内容收集整理于网络，其中内容若有侵犯您的权益，请留言告知；

2、本站概不提供真人lora模型，且本站模型仅为爱好者之间互相交流学习之用，严禁商业用途；

3、ckpt或safetensors结尾的主模型（2G以上）放在models/stable-diffusion文件夹里；lora模型（几十至几百M大小）放在models/lora，VAE放在models\VAE；

Shanzhagao 山楂糕 / Some sort of style Lora

admin 国风 2023-04-16 480 0

作者：ayyy1000022

触发词：hanfu，red lips，light rays，east asian architecture

masterpiece, best quality, extremely detailed, detailed background, detailed face, 1girl, solo, black hair, black eyes, blush, [straight hair|wavy hair], (messy hair, fluffy hair, long hair:1.1), scenery, atmospheric lighting, light rays, bokeh, depth of field, sparkle, red lips, child, hanfu, branch, flower, jewelry, crescent moon, light smile, wide sleeves, glowing

Negative prompt: EasyNegative, by (bad-artist:0.6), (low quality, worst quality:1.4), (monochrome:1.3), (bad anatomy), extra digit, fewer digits, (extra arms:1.2), bad hands, multiple girls, animal ears, fake animal ears, cat ears, fox ears, (multiple girls:1.4), furry, furry female, furry male, furry with non-furry, furry with furry, wide hips, thick thighs, huge ass, thick hips, cat girl, fox girl, animal girl, (realistic, clone:1.2), no humans, muscular, muscular male, (text, logo, artist logo:1.3), no humans, realistic, greyscale,

Size: 2048x3072, Seed: 1863413245, "mode: Horizontal, Model: Based64mix-V3, Steps: 24, Sampler: DPM++ 2M Karras, Use base: False, CFG scale: 14, Clip skip: 2, Mask blur: 4, Base ratio: 0.2, Model hash: fdf0096972, Use common: False, Use N-common: False", divide ratio: 1,1, AddNet Enabled: True, AddNet Model 1: shanzhagao128dim-epoch-000010(8d2dba01e9fd), AddNet Module 1: LoRA, AddNet Weight A 1: 1, AddNet Weight B 1: 1, Denoising strength: 0.3, Tiled Diffusion method: MultiDiffusion, Tiled Diffusion overlap: 48, Tiled Diffusion upscaler: 4x-UltraSharp, Tiled Diffusion batch size: 8, Tiled Diffusion tile width: 112, Tiled Diffusion tile height: 112, Tiled Diffusion scale factor: 2