大家好，我是chosen鲷，这次带来的是瑞典艺术家snatti笔下的小柯基lora。
snatti是一个典型的原画CG风格的画师，整体风格偏印象派，大的氛围感非常棒，尤其擅长刻画可爱温馨的小动物与自然景观一类的，我非常喜欢ta画的小柯基，一度还设置为我的微信头像，所以想要训练一个lora。
这个lora是专门出小柯基的，后面我还会做这个画师的lora。权重别整太高，0.7就行，不然容易崩坏。触发词：corgi style bysnatti
推荐的tag：
(masterpiece:1.2), best quality,PIXIV, corgi style by snatti,
这里写你想要的内容
<lora:corgi style by snatti_20231101164707-000014:0.7>
Negative prompt: EasyNegative, badhandsv5-neg,Subtitles,word,
Steps: 32, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 474290011, Size: 768x512, Model hash: 52b8ebbd5b, Model: chosen-mix bake vae, Denoising strength: 0.5, Clip skip: 2, Hires upscale: 2, Hires steps: 12, Hires upscaler: R-ESRGAN 4x+, Lora hashes: "corgi style by_20231101164707-000014: 4e2fbfae46ff", Version: v1.4.1
大模型我推荐用我的chosen mix（就是下方推荐里第一个），我的其他模型诸如chosen irises mix、impasto之类的也可以，其他的常见大模型例如ghostmeina之类也都均可，但是推荐用厚涂画风的二次元插画模型.
The following content is from Google Translate
Hello everyone, I am Chosen鲷, and this time I bring you Lora, the adorable Corgi drawn by Swedish artist Snatti.
Snatti is a typical concept CG-style artist, with an overall impressionist style. The atmospheric feeling in their artworks is amazing, especially when it comes to depicting cute and heartwarming animals and natural landscapes. I really love Snatti's drawings of Corgis and even had one as my WeChat avatar. So, I would like to train a model specifically for Lora.
This Lora is designed specifically for Corgis, and later on, I will also create a model for this artist. Please set the weight not too high, around 0.7, to avoid any instability. The trigger phrase is "Corgi style by Snatti."
Recommended tags:
(masterpiece:1.2), best quality,PIXIV, corgi style by snatti,
write anything you want here
<lora:corgi style by snatti_20231101164707-000014:0.7>
Negative prompt: EasyNegative, badhandsv5-neg,Subtitles,word,
Steps: 32, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 474290011, Size: 768x512, Model hash: 52b8ebbd5b, Model: chosen-mix bake vae, Denoising strength: 0.5, Clip skip: 2, Hires upscale: 2, Hires steps: 12, Hires upscaler: R-ESRGAN 4x+, Lora hashes: "corgi style by_20231101164707-000014: 4e2fbfae46ff", Version: v1.4.1
For the large model, I recommend using my Chosen Mix (the first one in the recommended models below). My other models, such as Chosen Irises Mix and Impasto, are also suitable. Other commonly used large models like Ghost Meina can also be used, but I recommend using a thick-paint, anime-style illustration model.