Sign In

cuteTMix2

104
648
6
Updated: Oct 5, 2024
base model
Verified:
PickleTensor
Type
Checkpoint Merge
Stats
648
0
Reviews
Published
Aug 2, 2023
Base Model
SD 1.5
Training
Steps: 20
Epochs: 10
Usage Tips
Clip Skip: 2
Hash
AutoV2
1E4E2296AE

感谢cuteYuki,如该模型作者有任何不满我将立即删除我的模型。

Thank you to cuteYuki. If the author of this model has any dissatisfaction, I will immediately delete my model.

The English version is below.

参数建议:Steps: 30, Sampler: DPM adaptive, CFG scale: 7-10,Denoising strength: 0.2,Hires upscaler: R-ESRGAN 4x+ Anime6B

本模型对比cuteTMix体型较接近萝莉,相应地,对cuteYuki面部优点保留的也较多,其实这个模型更可爱我用的更多。

本人最喜欢的模型之一cuteYuki对萝莉体型表现优秀,但是并不能很好的描绘少女,于是我手动采用多次分层融合变换身型,产出cuteTMix和cuteTMix2,本模型旨在保留cuteYuki可爱的面部,使其具有更成熟的体型。

使用少量提示词即可创造出一幅精美图片,过分提高场景描绘比例会使人物细节流失。

注:该模型会有肢体损失,乃正常情况,先抽抽卡吧,有时间我会修的,有时间的话

The following is translated by GPT:

Parameter suggestions: Steps: 30, Sampler: DPM adaptive, CFG scale: 7-10, Denoising strength: 0.2, Hires upscaler: R-ESRGAN 4x+ Anime6B.

This model compared to cuteTMix has a body shape closer to a loli, and accordingly, it retains more of cuteYuki's facial features. In fact, I find this model cuter, so I use it more often.

One of my favorite models, cuteYuki, performs excellently in depicting a loli body type, but it doesn't do well in portraying a young girl. Therefore, I manually used multiple layered fusion transformations to create cuteTMix and cuteTMix2. The purpose of this model is to preserve cuteYuki's adorable facial features while giving it a more mature body shape.

With just a few prompts, you can create a beautiful image; however, excessively increasing the scene depiction ratio may result in the loss of character details.

Note: This model may have limb losses, which is normal. Let's try it out first, and I will fix it when I have time. If I have time, that is.