WILL UPDATE INFO LATER

just trying to make a model that makes cool images :)

I like dark stuff so it might have some dark elements to it along with fantasy.

Version 2:

Tried to make things a bit better, better faces, a bit more color, and a bit more nsfw support

This model is a work in progress, I will try to updated it from time to time.

(let me know if you want me to add the baked version for V2)

recommend (but not needed)

easynegative - https://civitai.com/models/7808/easynegative

Deep Negative - https://civitai.com/models/4629/deep-negative-v1x

highly recommended

vae - https://huggingface.co/hakurei/waifu-diffusion-v1-4/blob/main/vae/kl-f8-anime2.ckpt

or

https://huggingface.co/stabilityai/sd-vae-ft-mse-original/blob/main/vae-ft-mse-840000-ema-pruned.ckpt (using this one currently)

upscaler - https://mega.nz/folder/zZhA1KoD#ds2nmgDNV4hfFpNSfEqN6Q -- 4x-UniScaleV2_moderate

Suggested negative tags: (worst quality, low quality, normal quality:1.7), lowres, normal quality, ((monochrome)), ((grayscale)), ((text, font, logo, copyright, watermark:1.6))

DPM++ 2M Karras works good

Hiresfix (step:15, denoise strength: 0.1-0.5)

Please use hiresfix or upscaling before inpainting. Results come out better for inpainting when you work on an upscaled image.

Resize mode, “Just resize (latent upscale)”

Mask blur at 3

Have inpaint area is set to “only masked”

Increase width and height to anything above 600

Set denoising strength to anything above 40 but below 55 (for faces)  - I use 48 as default and adjust if I have to

for a less realistic look and more anime looking check out my other model -- https://civitai.com/models/29578/jucyani666mix

Blonde slutty bimbo with huge tits that has a cock instead of a pussy

Mix Realistic Model

Lora

vae-ft-mse-840000-ema-pruned

https://civitai.com/models/29771/mix-anime-realistic-v15

Several people asked, so I decided to upload my custom mix. Small custom model, I mixed it for a long time, but it seems that it was grapefruitHentaimodel and AbyssOrangeMix3 by 0.5, and after underRealisticPornMerge by 0.25. The mix generates 2d art and nsfw well. Share your reviews and your works, it will be interesting for me to look at them))

this a general furry capable model

Visit Civitai-Crystal team on discord! -news, updates, development, beta testing

Join RunDiffusion on discord! - Auto1111 and InvokeAI in the cloud.
------------------------------------------------------------------------------------------------------
In the attempt to break the barrier between what's digital and what's real, one model rose to the challenge:

Crystal-Clear-Photorealistic

Now you can generate images as realistic in appearance as your prompt allows it.
This model was not developed for something specific so feel free to try out different things.
From regular portraits to sci-fi aliens in the 5th dimension. It should be able to generate a full body view of things at 704x1088(raw) without any mutations or duplicates.
You can go far, with your creations, by having a good prompt that has a good negative to it as well.

This model is the accumulated result of intense testing, training, feedback and minute detail improvements, upon the Crystal framework, done by an entire team of people over a 3 week period.

__________________________________________________________________________________________________

If your computer is not strong enough, a solution to your problem is RunDiffusion where you can generate your favorite CIVITAI models or you can train new models/LORAs or embeddings.

The model is based on the Crystal-Framework which was created to produce clean, noiseless and unrivaled image quality, while adjusting the colors and sharpness of the image. Although, CC-Photorealistic cannot compete with Crystal-Clear when it comes to producing proper content at very high resolutions, it is unrivaled in the realism domain. Be it food, houses, interiors, animals, nature, photography the choice is yours.
Future versions of this model will tackle the resolution issue !

Note: To produce high realism images, it is advise to use a low CFG ( between 4 and 7).

The model was not tested for mature content or how well it produces it !
Feel free to post your images and feedback, also, don't forget to join the Discord (extra goodies and models available):
Visit Civitai Crystal-Clear on Discord!

If you do plan to include this model in your creations, please show us what you have accomplished! We will also help promote your model if you give us the deserved credit.

This is a re-merge of Butterscotch that is intended to lean a bit more towards anime and produce images with better colors.

Capable of both male and female generations (leans more towards NSFW), though I never really generate many male centered images.

Future version may be implemented to help with hand generation (Hands are usually generated pretty well but not perfect).

No baked VAE in this version.

None of the sample images have used controlnet or been impainted to show how the model truly performs.

All sample images have been generated using Euler a and upscaled with foolhardy_Remacri at denoising strength 0.3.

Additionally the sample images were generated with no VAE being used.

Her name is RUIS. v1.2

The original RUIS model did not consider updates.

However, I found a direction to improve the model and took time to update it.

If there is a capacity problem, it is okay to use the existing version.

Added a distinct outer line and some NSFW.

I hope you enjoy using it.

I'll post the sample prompt as POST.

thank you,,,

Prompt

• (write down your imagination)

Negative

• (EasyNegative, badhandv4)

Default Setting.

• USE VAE "clearvae_main.safetensors"

• https://civitai.com/models/22354 with download

• Steps: 30,

• Sampler: DPM++ SDE Karras,

• CFG scale: 7,

• Denoising strength: 0.4,

• Clip skip: 2,

• Enable Dynamic thresholding enabled: True, Mimic scale: 7

• Script -> Detection Detailed -> bbox/mmdet ***

Step-by-step method of outputting as shown in the figure

First Step

• Hires. fix -> R-ESRGAN 4x+ Anime6B -> Hire Steps 10 -> Denoising strength: 0.4,

Second Step

• send to inpaint ->

• Paint both eyes of the character black with Mask Blur

• Enable Dynamic thresholding enabled: True, Mimic scale: 7

• Script -> SD upscale -> upscaler: 4x-UltraSharp,

To recap,

the first upscale is R-ESRGAN 4x+ Anime6B, the second upscale is 4x-UltraSharp after painting both eyes in inpaint. Upscale a total of 2 times to improve the quality.

Back to the origins of All Things That Is.

NAI full, Anything V3 full, Counterfeit 2 full, F222 (very small weight).

OriginMixFull.ckpt - for merges and training, OriginMix.safetensors - for inference.

《游戏图标研究所》

This is a model of a large test game ICON, which belongs to the beta version. Welcome to join the QQ group: 489141941 to discuss and give comments, which is convenient for model development. This model is mainly suitable for the generation of game icons. There is no need to add back tags when using it. This version is mass-produced by the Icon Academy model, and everyone is welcome to use and comment for better optimization in the future.

1. Icon 2.0 (official) version will be released on May 1

2. Contains French style, two-dimensional and European and American cartoons

3. Fix the poor quality of version 1.0, the problem of being uncontrollable, and solve the problem of generating strange pictures

4. After the release, the test will start in the QQ group

Project All In One (AIO)

Please Red Description

Important: Having multiple models uploaded here on civitai has made it difficult for me to respond to each and every comment. One of the ways I plan on addressing this is via the creation of a pdf guide for each and every model (think of it as the models documentation). This will take a while though. So in the meantime, if you have any questions or feedback -

visit my thread of the Unstable Diffusion Discord

This model is a merge between all of my previously released models and, some unreleased models.

Going forward, all future models and updates to existing models will also be added/updated on this model.

Hands

Hands are the Achilles heel of latent diffusion models (LDMs). This model is no exception. Due to the unorthodox methods used in the creation of this model. It often fails at generating good hands. As of right now, I have no fix for this. Given the dynamic nature of this model (being updated alongside my other models), there's a chance the issue solves itself in a later update. That being said, I also plan on dedicating time into researching possible cost-effective solutions to "the hand issue" for all LDMs.

In the meantime, I encourage you to use the following two embeddings to help alleviate some of the deformed hand generations:

• Bad_prompt_version2 - https://huggingface.co/datasets/Nerfgun3/bad_prompt/blob/main/bad_prompt_version2.pt

• bad-hands-5 - https://huggingface.co/MortalSage/Strange_Dedication/blob/main/embeddings/bad-hands-5.pt

NOTE: If using bad-hands-5 bare in mind that your generations will not match the sample images for this model. This is because I didn't use the bad-hands embedding when creating the sample images.

How To Install Embeddings in Automatic1111 webui

Place bad_prompt_version2.pt and/or bad-hands-5.pt inside your embeddings folder for webui.

stable-diffusion-webui//embeddings//<place .pt embeddings here>

Launch webui, under "Generate" you will see a little red button. Click it, now under the "textual inversions" tab you will see the embeddings you added. Click on them, they will automatically be added to the end of your negative prompt.

To change the weight (strength) of the embedding; use attention/emphasis. For example, (bad_prompt_version2:0.8)

Recommended Settings

Clipskip: 1

ENSD: 31337

Sampler:

• DPM++ SDE Karras, 18 - 30 steps

• DPM++ 2M Karras, 30 - 60 steps

• Heun, 20 steps, Sigma Churn = 1

• Euler, 20 - 70 steps, Sigma Churn = 1

These parameters are not strictly required, experiment around with other samplers and parameter values. You might find something that works better for you.

Check out my other models (also the models used in the merge)

• Eris - https://civitai.com/models/21952/eris

• WonderMix - https://civitai.com/models/15666/wondermix

• Refined - https://civitai.com/models/8392/refined

• Experience - https://civitai.com/models/5952/experience

• Elegance - https://civitai.com/models/5564/elegance

• Clarity - https://civitai.com/models/5062/clarity

• VisionGen - Realism Reborn -https://civitai.com/models/4834/visiongen-realism

LoRA

• Pant Pull Down - https://civitai.com/models/11126/pant-pull-down-lora

Special thanks to rocp for the witch and French maid prompt used in the sample pictures, @nutrition for copious research/testing of various sampling methods/parameters, and to everyone in the Unstable Diffusion Discord community that make doing this a more fun and enjoyable experience.

This model is mixed from mixpro4, Dark Sushi Mix, majicMix fantacy and some of the other models. With an anime style character and detailed(I think so) background,

You can use the vae-ft-mse-840000-ema-pruned and vae_校长滴二次元school anime as VAE, the later has a higher saturation.

For the exemple pictures, the name in the information is different to the model name, that was for me to remember the basic models.

Some of the generation information are deleted while updating, because it might violate the TOS.

融合了一些自己喜欢的模型

2023.5.1 GhostMix-V1.2

UPDATE DETAIL(中文更新说明在下面)

Please Give Me 5 Stars if you like my model. Thanks a lot.

GhostMix V1.2 is an absolutely astonishing model, and I think it is the strongest 2.5D model in Civitai right now. I think a update of the model should improve the model’s compatibility, good image rate and image details given the main structure of 90% generated images doesn’t change. So I use layer combination to combine models layer by layer. And I got this model after 9 different versions of “the final version of GhostMix V1.2”,LOL.This version of GhostMix V1.2 is a balance in terms of compatibility, good image rate and image details, although sometimes using the same Promts of GhostMix V1.1, may comes up a different result.But this doesn’t happen too much. I already tried my best, I hope you guys like it.

PS: The first image of GhostMix V1.2 is a nod to Mamoru Oshii’s version of Ghost in Shell in 1995, and my name Ghost_Shell is also a nod to this movie.

中文更新说明

GhostMix V1.2是绝对让你惊艳的模型，也是自己认为现在最强的2.5D模型。我认为模型的更新应该是基于现有的画面整体不大变的前提下，提高模型的成图率，兼容性和画面细节。所以我采用了分层融合，一共做了9个版本的“GhostMix V1.2最终版本”，最终得到了现在这个版本的GhostMix V1.2。这个版本的GhostMix V1.2在兼容性，成图率和画面细节表现比较平衡，虽然有时候会出现同样的Promts，出图改变的情况，但实测不多。我尽力了，希望大家喜欢。

PS：GhostMixV1.2的头图是致敬1995押井守版Ghost in Shell 攻壳机动队的生成的，Ghost in Shell也是作者名字的由来。

2023.3.11 GhostMix-V1.0

Introduction(中文简介在下面)

First of all , I want to thank all the people who use this Checkpoint. And this is my first Checkpoint.All the sample image can be reproduced. This Checkpoint works well on both SFW and NSFW.THE NSFW PART IS VERY GOOD!!!
I uploaded this Checkpoint yesterday and from yesterday to today , I am still trying all the possibility of this Checkpoint. So if you try the model and find some good promts , I hope you can upload it and share with me, it will help me for the next version of GhostMix, Thank you agian!

Recommend Some Promts:

1. Fractal Art（highly recommend，awsome）

(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), (1girl:1.3), (fractal art:1.3),

1. Realistic Art-Girl & Flower field

((masterpiece, best quality)), 1girl, flower, solo, dress, holding, sky, cloud, hat, outdoors, bangs, bouquet, rose, expressionless, blush, pink hair, flower field, red flower, pink eyes, white dress, looking at viewer, midium hair, holding flower, small breasts, red rose, holding bouquet, sun hat, white headwear, depth of field,

1. Color Art

(flat color:1.3),(colorful:1.3),(masterpiece:1.2), best quality, masterpiece, original, extremely detailed wallpaper, looking at viewer,1girl,solo,floating colorful water

1. Realistic Art-Nude Girl(NSFW-recommend)

(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.3),  illustration, beautiful detailed eyes, (1girl), (nude),detailed nipples,huge breasts,puffy nipples, oniomania, erotic,beautiful detailed pussy,soft breasts,upper body

VAE&Textual Inversion:

VAE: kl-f8-anime2 or vae-ft-mse-840000-ema-pruned(anime suggest: kl-f8-anime2)

Textual Inversion: ng_deepnegative_v1_75t, easynegative

About Image Reproduction:

Some user said that they can not reproduce my result.Maybe the setting goes wrong.I just reproduce my cover image. If you want to reproduce my result, Checkpoint Model，Postive Promt，Negative Promt，Textual Inversion，Sampler Steps，Sampler，CFG，Resolution，Seed , ALL the things should be the SAME! Then be careful if you open controlnet or lora ,don't make them influnence your result.

中文简介

首先感谢每一个使用这个Checkpoint的人，这是我融的第一个Checkpoint。所有样图自测都可以复现。SFW和NSFW的图都挺漂亮的，NSFW的图非常棒。这个模型昨天上传到今天，我一直在尝试这个模型的可能性。希望使用这个Checkpoint的朋友们，如果你试了，觉得有不错Promts，欢迎上传图分享，让我也了解一下这个Checkpoint的可能性，也可以帮助我调下一个版本GhostMix，再次感谢。

推荐Promts:

1. 分型艺术（超级推荐，必出好图）

(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.2), (1girl:1.3), (fractal art:1.3),

1. 真实风格-少女与花海

((masterpiece, best quality)), 1girl, flower, solo, dress, holding, sky, cloud, hat, outdoors, bangs, bouquet, rose, expressionless, blush, pink hair, flower field, red flower, pink eyes, white dress, looking at viewer, midium hair, holding flower, small breasts, red rose, holding bouquet, sun hat, white headwear, depth of field,

1. 色彩艺术

(flat color:1.3),(colorful:1.3),(masterpiece:1.2), best quality, masterpiece, original, extremely detailed wallpaper, looking at viewer,1girl,solo,floating colorful water

1. 裸女(色图-推荐)

(masterpiece, top quality, best quality, official art, beautiful and aesthetic:1.3),  illustration, beautiful detailed eyes, (1girl), (nude),detailed nipples,huge breasts,puffy nipples, oniomania, erotic,beautiful detailed pussy,soft breasts,upper body

VAE&Textual Inversion:

VAE:kl-f8-anime2或者vae-ft-mse-840000-ema-pruned(动画风建议kl-f8-anime2)

Textual Inversion:ng_deepnegative_v1_75t,easynegative

关于图片复现:

有同学好像没法复现我的图，可能是设置有点问题，刚刚我才把头图给复现了。注意几个点：Checkpoint 模型，正向Promt，反向Promt，Textual Inversion，迭代步数，迭代方法，CFG，分辨率，随机种子都必须一模一样！然后复现的话，要把controlnet和lora关掉，怕影响复现。

[[Basically, I'll give you a heads up that I didn't do any (NSFW) testing. (NUDE is possible) If you want (NSFW), please merge your models separately.]]

[[[[recommend using Clipskip 2]]]] <- Never ignore it

Introducing Cutmix, a model that competes solely on cuteness. I won't write much about its usage, as I don't think models need much explanation these days.

Basically, it reflects Lora in a good way.

Her fingers don't twist like crazy, but they don't come out too well either.

The downside is that it's basically a circular model, which is a shame if you want a more mature look.

After you get the look you want, use an upscaler like Tiled VAE or sd upscale to get a better quality photo (smaller upscalers like ddetailer will work too).

Also, be forewarned that this is a heavy model. We used a model converter to make it as light as possible, but we warn you that it may be difficult to use if you're using Colab or a low-end graphics card (for fp32).

Negative Prompt: rMadaNegative4 (SD 1.5) | Stable Diffusion TextualInversion | Civitai

VAE: stabilityai/sd-vae-ft-mse-original at main (huggingface.co)

Optional: rMadArt3_NoiseOffset (SD 1.5 Lora) | Stable Diffusion LORA | Civitai

Anime Pastel Dream

CHECK BOTH VERSIONS BELOW

Add a ❤️ to receive future updates.
Do you like what I do? Consider supporting me on Patreon 🅿️ or feel free to buy me a coffee ☕

Available on Sinkin.ai with GPU acceleration.
MY MODELS WILL ALWAYS BE FREE.

This is my attempt at making a good anime pastel model. I love the end result and I hope you enjoy it. This should also work great with most anime models.

I made 2 versions:

• Hard has a stronger pastel style and gets a lot more creative.

• Soft has a subtler pastel style and is much easier to control.

Consider supporting model creators of the models you like!

Suggested settings:

• CLIP skip 2

• I had ENSD: 31337 but it's optional

• NO restore faces

• Highres fix with either a general upscaler and low denoise or Latent with high denoise (see examples)

• Be sure to use Auto as vae for baked vae versions and a good vae for the no vae ones. Avoid anythingv3 vae as it makes everything grey.

• Steps and upscale denoise depend on your samplers and upscaler. See the examples.