VAE

I dunno what to write ¯\_(ツ)_/¯ but i will give you some tipsV2.0 UpdateChanged some recipe the 2.0 model will lean into anime style/kemono style more than western style but i...

3D + 2D + hairstyles + clothing + photographyImages should have magazine proportions i added this dimensions to all training dataset:512x768768x1152>>>post more portra...

QQ群: 816567536 欢迎来玩，交流，找我做LORA,This is an anime style model with a certain tendency towards more big breasts and sexiness in character output, but not too much to d...

描述: 训练词语: 名称: clearvae_variant.safetensors ...

Works best between 0.8 & 1.0All my models are free, but if you would like, you can support my work via Buy Me a Coffee or Ko-fi!Boogiepop prompt:touka miyashita, hat, cloak...

该系列模型为梦想家系列，主要为我个人喜欢的可爱画风的收藏，主要来源与对novelai系列模型画师组合的迁移，不限于模型的类型与风格，目前已经开始积极尝试XL版本的cuteyukimix可爱模型，该...

Model introduction: This is a character street photography as the theme of the model, the maximum size of the data set is 1232px, should be clear enough, training deployment 90...

Use CPU as RNG in Automatic1111/Forge or just ComfyUI. ComfyUI is best choice. All the img in preview are created in ComfyUI.Model Overview:The Neuro-Evolution:I've been dedica...

Word of the AuthorFinally a model merge I got exciting enough for to post here after a long time of lurking and trying out different models, prompt styles and getting to know A...

Use CPU as RNG in Automatic1111/Forge or just ComfyUI. ComfyUI is best choice. All the img in preview are created in ComfyUI.Model Overview:The Neuro-Evolution:I've been dedica...

Help support me on Patreon! https://patreon.com/LightchopioNow accepting commissions! https://www.fiverr.com/s2/3403793c68Recommend high-res and strength of 0.8All images using...

Works best between 0.8 & 1.0All my models are free, but if you would like, you can support my work via Buy Me a Coffee or Ko-fi! 描述: ...

This was actually a funny model, because it started out as a model that Duskfallcrew was going to build on and make "DIGITAL FANTASY" - but something broke, and they shared the...

yes 描述: yes 训练词语: 名称: kl-f8-anime2_fp16.safet...

What This IsIt is a sloppy merge of Earthnicity's Holiday Mix and Analog Madness into Dreamshaper 8 using ComfyUI so I am not able to prune it. I tested it about 3 times to mak...

A simple checkpoint mix with a bias towards semi-realism style with the addition of a hand-drawn style.It is recommended to use VAE to enhance the image clarity ...

This model is just a merge of two wonderful models, it's not even close to the originals, but I ended up liking the results, and decided to share it.Recommended settings:Sample...

The purpose of this VAE model is to reproduce the grey tones of SD1.5 models that damage VAE这个是专门用于再SDXL复刻SD1.5烂VAE灰色调的VAE，你可以将其丢到模型里替换模型原有的VAE...

Version 1More details in the faceMore realistic representation of peopleBetter incidence of light on the bodybetter representation of the handstrained size 768x768, 1024x1024 ...

What This IsIt is a sloppy merge of Earthnicity's Holiday Mix and Analog Madness into Dreamshaper 8 using ComfyUI so I am not able to prune it. I tested it about 3 times to mak...