<h3>Model</h3><p>同时使用ti和lora你能在绝大部分模型中都得到不错的效果。</p><p>Using ti and lora together you can get good results in most models.</p><p></p><h3><strong>Why ti+lora?</strong></h3><p>ti和lora一起用能在照片基础的模型上工作的更好，而且不用写一大堆tag。</p><p>有时候你不能加太多人物lora的权重，就需要人物ti来补充人物特征。</p><p>Use both TI and Lora works very well with photrealistic models,and you don't neet to write a lot of tags.</p><p>Sometimes you can't add too much weight to the character lora, so you need character ti to complement the character characteristics.</p><p></p><p>lora+tags:</p><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/5e1dc2e9-9fd4-44da-3475-035c05593500/width=525/5e1dc2e9-9fd4-44da-3475-035c05593500" /><p>ti+lora:</p><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/b7def286-05b9-4996-59bd-22cc6581c200/width=525/b7def286-05b9-4996-59bd-22cc6581c200" /><p></p><h3>Use <a target="_blank" rel="ugc" href="https://github.com/hako-mikan/sd-webui-lora-block-weight.git">lora block weight</a> to limit lora</h3><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/24440ed1-cf07-4f64-4da3-e4f664767000/width=525/24440ed1-cf07-4f64-4da3-e4f664767000" /><p></p><h3>Problems:</h3><p>树王的图好少！由于图不足，当初训练的时候ti只给了12个向量，现在会有不小的概率生成出四只耳朵或者两根辫子，或者叶子头饰的位置不对。</p><p>Rukkhadevata's picture is so few! Due to a lack of graphics,I trained ti I only use 12 vectors.Sometimes well give four ears or two braids, or a leaf ornaments in the wrong place.</p>

Files with SHA256: bd6adbea488b4a8bf44ebbaea2b4cb9d86f5bcc0fa01541def58116549d255f7