This model is designed to generate Chibi_Emoji style lora.

此模型旨在生成Chibi_Emoji画风的lora，

Trigger Words：Chibi,50g

角色触发词 ：Chibi,50g

Lora weight coefficient： 0.5-0.9

• Lora权重： 0.5-0.9

When using this model, please carefully read the recommended setting parameters and model description in the introduction.

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

Ai绘画交流q群：813799800

Model description:模型说明：

This Lora has three versions [000008]

此Lora有1个版本[000008] ,

The following is the detailed weight test chart of the Lora version：

下面是Lora版本的详细权重测试图：

Recommended parameters:推荐参数：

Sampler：

• Euler a

• Heun

• DPM++ 2M

• DPM++ 2M SDE Heun Karras

• DPM++ 2M SDE Heun Exponential

• DPM++ 3M SDE Karras

• DPM++ 3M SDE Exponential

Steps: 28-50

Denoising strength: 0.5-0.6 【Please do not use “latent”】

Upscale by: 1.5+

CFG scale: 5-13

Prompt： under （写在最后）

• , best quality, amazing quality, very aesthetic, absurdres

Negative Prompt：

• ????, 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 ???????, bad hands, @_@, mismatched pupils, heart-shaped pupils, glowing eyes

width*high:

• 1:1= 512*512 , 640*640 768*768 , 869*869 , 1024*1024

• 2:3= 512*768 , 640*960 , 768*1024

• 16:9= 910*512 , 640*1138

• 1.85:1= 1184*640

• 2.39:1= 1224*512

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:https://www.buymeacoffee.com/EEEMax

如果您想支持我：https://www.buymeacoffee.com/EEEMax