This model is designed to generate characters such as lvzi（緑子）and the style of lora. I like the character lvzi（緑子） very much so I trained this model.此模型旨在生成lvzi（緑子）等角色以及画风的lora，我很喜欢lvzi（緑子）这个角色所以我训练了这个模型。<h2 id="heading-68">Lora weight coefficient：</h2><ul><li>Lora权重： 0.6-0.7-0.8-0.9</li></ul><h2 id="heading-69">Trigger Words：author,alter,lvzi</h2>角色触发词 ：author,alter,lvzi<h2 id="heading-10">Please use this lora with pony series models</h2><h2 id="heading-11">请搭配pony系列模型来使用本lora</h2><h1 id="heading-5204">When using this model, please carefully read the recommended setting parameters and model description in the introduction.</h1><h1 id="heading-5205">在使用此模型时请认真看简介中的推荐设置参数和模型说明</h1>Ai绘画交流q群：813799800<h2 id="heading-5206">Model description:模型说明：</h2>This Lora has three versions [000002], [000006], [1]. The training and learning levels of the model range from shallow to deep. Please choose according to the cover image of each version. (Each version will be uploaded one by one. Currently, only the cover image of [00006] has been completed)此Lora有3个版本[000005] , [000010], [1]。其分别是模型的训练学习程度由浅到深，根据各版本的封面图来请自行选择。（各版本会逐一上传）<h3 id="heading-5207">Below is the detailed xyz test chart of the Lora version:</h3><h3 id="heading-5208">下面是Lora版本的详细xyz测试图：</h3><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/337db880-2121-4def-a7fc-0711520fb2d4/width=525/337db880-2121-4def-a7fc-0711520fb2d4.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1a1e4bea-ca20-47cc-ac1a-963b8af1fcff/width=525/1a1e4bea-ca20-47cc-ac1a-963b8af1fcff.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6ab5ef4c-c9ba-4309-af9e-00dc2a49841b/width=525/6ab5ef4c-c9ba-4309-af9e-00dc2a49841b.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4c2ccd3b-8933-4cf7-b2c9-6a7986512217/width=525/4c2ccd3b-8933-4cf7-b2c9-6a7986512217.jpeg" /><h2 id="heading-5209">Recommended parameters:推荐参数：</h2>Sampler：<ul><li>Euler a</li><li>Heun</li><li>DPM++ 2M</li><li>DPM++ 2M SDE Heun Karras</li><li>DPM++ 2M SDE Heun Exponential</li><li>DPM++ 3M SDE Karras</li><li>DPM++ 3M SDE Exponential</li></ul>Steps: 28-50Denoising strength: 0.5-0.6 【Please do not use “latent”】Upscale by: 1.5+CFG scale: 5-13Prompt： <pre><code>score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up,</code></pre>Negative Prompt：<ul><li>nsfw, lowres, {bad}, error, fewer, extra, missing, worst quality, jpeg artifacts, bad quality, watermark, unfinished, displeasing, chromatic aberration, signature, extra digits, artistic error, username, scan, [abstract], bad anatomy, bad hands, @_@, mismatched pupils, heart-shaped pupils, glowing eyes</li></ul>width*high:<ul><li>1:1= 640*640 , 768*768 , 869*869 , 1024*1024</li><li>2:3= 512*768 , 640*960 , 768*1024.....</li><li>16:9= 910*512 , 640*1138 , 768*1365</li><li>1.85:1= 1184*640</li><li>2.39:1= 1224*512</li></ul>Please take the generation of prompt words seriously. Only in this way will produce better results请认真对待生成提示词只有这样才会产生出更好的效果Note Do not use negative embedding when there is no problem with the image.注意 当图像没有问题时，请勿使用负嵌入。Credit me if you use my model in your own merges如果您在自己的合并中使用我的模型，请注明出处Please comment if you have any questions.如有任何疑问，请发表评论。if you would like to support me:<a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/EEEMax">https://www.buymeacoffee.com/EEEMax</a>如果您想支持我：<a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/EEEMax">https://www.buymeacoffee.com/EEEMax</a>

Files with SHA256: 2f656a9739b1b5e9c042f8cc4e44a956f1c0279b20780b561af83bb906d93bdd