This is an evolution of my model (base mix[jobtang mix]). I've tightened up the colors throughout, and bumped up the hand batting average a bit. I reduced the strong pastel tones and strengthened the background.

You can use it as you normally would, it's totally fine. Sampling is DPM++ SDE Karras, DPM++ 2m karras. Upscalers are AnimeSharp or R-ESRGAN 4x+ Anime6B.

Sampling step is recommended to be 30 over

You can use Dynamic Thresholding to get more interesting colors.

++Added a lightweight model for FP16!

私のモデル（base mix[jobtang mix]の進化版です。全体的に色味を引き締め、手の打率を少し上げました。パステル調の強い色を減らし、背景を強くしてみました。

普通に使っても全く問題ないのですが、それでも一番良いものをオススメするなら、サンプリングはDPM++ SDE Karras、DPM++ 2m karrasを使うことをオススメします。 アップスケーラーはAnimeSharpまたはR-ESRGAN 4x+ Anime6Bです。

サンプリングステップは30オーバーを推奨します。

Dynamic Thresholdingを使用すると、より面白い色を得ることができます。

++FP16用の軽量モデルを追加しました！

If you like the model, help me with some of the electricity bills -> https://yuribun500.fanbox.cc/

If you have any questions, please contact us via Pixiv or Twitter -> https://twitter.com/nzxt_cam

I share the model that I used to make the Futanari model.

I wanted to implement between a flat anime feeling and a 3D picture body through model merging, and this is the result.

If you want an NSFW picture, you should highlight it using parentheses. Also, I recommend using LoRA for various sex postures.

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.

PurePerfection is just a mix of my favorite models, this is my go to for realism. I can't recall which models off the top of my head. Model works decent in NSFW environments, could be better. Poor prompting won't get you far with this mix, and please remember to use a VAE. vae-ft-mse-840000-ema-pruned

小插画。太懒了不想写任何介绍。

总之这是一个画风如图所示的模型

建议开高清修复，开一点点就能得到更好的效果

推荐的原始图参数

Negative prompt:(bad and mutated hands:1.3), (worst quality:1.5), (low quality:1.5), (blurry:1.5),

Steps: 10, Sampler: Euler a,CFG scale: 7,Size: 512x640,Clip skip: 2,

推荐的高清修复参数(放大倍率1.2以上都会有比较好的效果)

R-ESRGAN 4x+,steps:10,Denoising:0.5,

ღ( ´･ᴗ･` )比心

以下机翻

little illustration. I'm too lazy to write any introduction.

It is recommended to open HD repair, open a little can get better results

Recommended raw graph parameters

Negative prompt:(bad and mutated hands:1.3), (worst quality:1.5), (low quality:1.5), (blurry:1.5),

Steps: 10, Sampler: Euler a,CFG scale: 7,Size: 512x640,Clip skip: 2,

Recommended HD repair parameters (magnification above 1.2 will have a better effect)

R-ESRGAN 4x+,steps:10,Denoising:0.5,

ღ( ´･ᴗ･` )比心

I'm no naming genius.

Photoreal is just marketing bs, but yeah, the texture is pretty good. I know it's an oversaturated model type, but hopefully this one has some qualities you'll enjoy.

The product of haphazard experimental alchemy with Supermerger, the outcome was a lot better than I expected. The goals going in pretty much came through, sheer blind luck. It's a similar premise to Ares Mix's idea of using an anime core to expand pose repertoire; as well as the dramatic dark lighting capacity from Experience.

Experience Realistic / Ares Mix / Deliberate

Liberty / URPM / Realistic Vision

Yesmix / Sparkling Nectar / Perfect World

I highly recommend this negative embedding. It has an extremely powerful effect on outputs, see the examples. From what I've seen it usually kills the ability to have dark images unless you use [x:n.n] to keep it off for a certain number of steps, which allows the model to set the scene. And it loves freckles, so be sure to neg that out if needed.

I initially didn't think this needed a VAE, most the example images don't have one active. The blonde lady and embedding on/off compares do. It's pretty acceptable without, but use the good ol' vae-ft-mse-840000-ema-pruned to kick it up a notch.

I'll be adding the intermediate constituents of the mix to this post at some point, you can pop them through Supermerger to play with them or whatever. Mainly I just need to rescue some disk space.

If the anime model looks interesting, it'll be up soon. I was going to stick it here but maybe it's worth its own post.

Generates pinup models, clothed or unclothed

I trained 10 LoRa's on 1300 realistic photos of very high quality at 1024x1024 resolution and mixed them with the Holygenex model, then combined the resulting model with the SD 1.5 base model.

The resulting Holygenex LR model can generate 768x768, 1024x768, 768x1024, even 1024x1024 if hardware allows (requires more than 12GB of video memory to use Hires. fix)

The task is to create a model which can generate high resolution images with excellent quality on the SD 1.5 engine.

From now on, you can say that SD 1.5 can generate images with 768x768 and 1024x1024 resolution.

To get a good quality image you need to use Hires. fix. There is no other way!

The images in the examples are generated at 100-150 steps and Hires. Fix with Denoising strength: 0.3 using Upscale by 2 with best upscaler 4x-UniScale-Balanced [72000g]. See the full picture descriptions in the examples. Click "Copy generation data", copy the text to notepad and see the full description.

Please don't write that you don't get good pictures in 512x512 resolution at 20 steps. This model is not for small resolutions. This is a beta version of the model to create images in large resolutions, so you will get crumpled results on small pictures.

In the beta version you can generate lanscapes, portraits of women and men, cars, motos, robots, interiors and streets.

For best results use the VAE, vae-ft-mse-840000-ema-pruned https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.ckpt

This embeddings must be used in your NEGATIVE prompt:

bad-hands-5 https://huggingface.co/nolanaatama/embeddings/blob/main/bad-hands-5.pt

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

Negative Prompt for example:

EasyNegative, bad-hands-5, (((nude, naked,child, child face))), un-detailed skin, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, ugly eyes, (out of frame:1.3), worst quality, low quality, jpeg artifacts, cgi, sketch, cartoon, drawing, (out of frame:1.1), worst quality, low quality, jpeg artifacts, poorly drawn, (((word, words, letter, text, signature, watermark)))

Animating very pretty shapes

More lightweight, resulting in great raytracing and CG quality at low stabs and CFGs.

Prompt

(masterpiece:1.2, best quality:1.1), full body,

Prompt + Ray Tracing, Global Illumination,
Feel free to add more using the Styles feature.

An Extremely Delicate And Beautiful, Best Shadows, Shallow Depth of Field, Brilliant Colorful Paintings, Dramatic Shadow, Dramatic Lighting, sharp focus, (8k), (4k), extremely detailed, intricate, hyper detailed,

Negative prompt

(worst quality:1.4), (low quality:1.4), multiple panels, blurry, bad-hands-5, bad_prompt_version2, multiple panels, blurry, watermark, letterbox, text, ( Poorly Made Bad 3D, Lousy Bad Realistic:1.1 ), From Above, From Below, multiple views:1.4,

- Negative prompt /( From Above, From Below )/ You can leave it out.

Steps : 30 ~ 45 / max 55

CFG : 6~13 / max 22.5
Denoising strength between 0.3 ~ 0.5
i'm love 0.4 and 0.5
Clip skip : 2



Purple Eyes -Settings

Detection Detailer - bbox\mmdet_anime-face_yolov3.pth

Padding value (in pixels) when inpainting at full resolution 3

CFG 8~15
I'm CFG 15 love

Enable Dynamic Thresholding (CFG Scale Fix)

Top percentile of latents to clamp 93~96
top Half Cosine Up
3

Below Half Cosine Up
3

Good luck with your images! :D

We'll be back with more great models!

Have a great day.

I wasn't too busy, so I remade Galena as well.

This is an NSFW prone model. It is meant to generate primarily women and NSFW scenarios. This was made with the new base model I merged together, alongside several style LoRA's I re-did myself.

I use Hires Fix with 4x Fatal Anime 50000 G at 15 Hires steps and 0.45 Denoising strength, upscaled by 2 times. I use the Orangemix VAE. Everything else is in the tags.

I also have a Discord. Join if you wanna share results, get advice, or make requests.

Upscaler: R-ESRGAN 4x+

Hires steps:26

Denoising strength: 0,7

Realistic and Anime! what if i can do! dragon ball, Sailor moon, anime.

▼ Merge Based On ▼

• ChilloutMix (Ni Pruned FP32 Fix)

• LOFI (2.1)

• 饭特稀 (0.8a)

• majicMIX realistic (4.0)

• Lyriel (1.5)

• BRA (Beautiful Realistic Asians) (4.0)

• ReV Animated (1.2.2)

▼ Recommended Settings ▼

Sampling Method

• DPM++ SDE Karras

• Euler a

• DPM++ 2S a

• DPM2 a Karras

Sampling Steps

• 40 (20 to 60)

CFG Scale

• 7 (6 to 9)

Clip Skip

• 1 (1 to 2)

VAE

• None (Baked)

________________________________________

Any questions or suggestions you can contact me through Discord Cisney Gassai#5108 or Twitter @Realisian.

For donations or requests you can do it on my Ko-fi page <3.

• https://ko-fi.com/cisneygassai

"Aira Lumina" is a huge additive model merge, as can be seen by its ~4gb fp16 size. It carefully combines well over 200 models, including both realistic and anime checkpoints.

This has been a labor of love, often doing and undoing merges based on the results. Probably the most impressive thing is the integrity of the results -- you have less control over whether you get anime/photorealistic photos, but it passes alexds9's overfit/corruption test with flying colors.

Because the merge is mostly additive, it performs well at both clipskip 1 and 2, and flipping back and forth can yield interesting results.

Enjoy!

The model name is an homage to the great Frank Frazetta. I made it to add in the style of some of my favorite artists (including Frazetta) I felt were not trained well enough in SD 1.5, or that weren't in it at all. All artists I have included have already passed away. Many come from the great pulp era of American publishing. I started with the same images used for another model, which the creator gave me and said that he won't be working on his any more. Then I cleaned up and edited all the text files, then added several more artists, adding a couple of hundred more training images in all. I haven't magically made hands always work or anything like that - you'll still need to inpaint for that. And I have used no inpainting for the sample images. I have simply cherrypicked a few I thought were good results.

Using the model should be pretty simple. Using the keyword "painting" will pull pretty hard to a mixed style. Other than that, you can use the artist names. Using "Fake Fauxzetta" won't do you any good.

• Allen Anderson

• Earle Bergey

• Frank Frazetta

• George Gross

• Gil Elvgren

• Glen Orbik

• Robert Maguire

• Robert Stanley

• Tom Lovell

Also, using keywords for pulp magazine genres will halp: western, scifi, fantasy, mystery, crime, romance...

Most of the description texts were in the following format:

a painting, by {artist}; a {subject}, {description of what the subject is wearing/doing}; a {another subject}, {description of what that subject is wearing/doing}; ... {info about the background or setting}; {genre tag and other important keywords}

So formatting your prompts like that will probably help yeild decent results.

For example:

• a painting, by Robert Maguire; a boxer, standing in the ring; dramatic lighting, action

• a painting, by George Gross; a woman, sitting on a motorcycle; at night, in the city, masterpiece, HQ

This model is trained at 512x512, and based on SD 1.5. I can only train at 512x512, as my video card is not a screaming beast. Maybe someday I'll take donations, and with them buy a nicer card and be able to produce a SD 2.1 version, and release my higher-res data set.

I don't really care what you do with this model, as long as it's legal. I just want to give something back to this community.

I used a set of instructions I found somewhere to make an inpainting version of the model, too. See the tab at the top of the model.

If you think you can do better than me at training a model, you are welcome to the data set I used. https://www.mediafire.com/file/9tig5ha7b50i1to/FakeFauxzetta-data-set.zip/file

This is a Checkpoint i merged from my 3 favorite Checkpoints. Dark Sushi Mix, ReV Animated, Pika's Animated Mix.

I have been getting some incredible results with this merge and thought that it would be worth sharing with others. It is great at generating both NSFW and SFW content. Would love to see your results :D

Check my other models if you liked this one :)

My recommendations is as follows:

Sampling Method: Euler A or DPM++ SDE Karras

(My personal favorite is Euler A)

Sampling steps: 20-30

(I use 50 for txt2img 512x512 and turn it down to 20 for img2img and increase the resolution to 1024x1024)

CFG Scale: 6-8

CLIP: 2

Hires.fix: upscale by: 1.5x -- Hires steps: 10 -- Denoising strength: 0.3

(I don't really use Hires.fix since i can generate 1024x1024 with my GPU, but if you can't then its recommended)

VAE: I strongly recommend to use this model with vae-ft-mse-840000-ema-pruned

For a more realistic image use:

photorealistic, (hyperrealistic:1.2), beautiful, masterpiece, best quality, extremely detailed face, perfect lighting,

For less realism use:

beautiful, masterpiece, best quality, extremely detailed face, perfect lighting,

My favorite negative prompt:

(worst quality, low quality:1.4), (monochrome), zombie, watermark, username,patreon username, patreon logo,

high quality anime style model.

more info. https://huggingface.co/gsdf/Counterfeit-V2.0

Verson2.5 https://huggingface.co/gsdf/Counterfeit-V2.5

Verson3.0 https://huggingface.co/gsdf/Counterfeit-V3.0

(Use clip: openai/clip-vit-large-patch14-336)
EasyNegative(Negative Embedding) https://huggingface.co/datasets/gsdf/EasyNegative

Please consider Buying Me A Coffee if you like my contributions to the community: https://www.buymeacoffee.com/Rjox

Differences between versions:

v5 - Preferred version - Flexible with more variety of faces, bodies and settings.

v6 - Less flexible and more of a specific look to the faces and bodies. It's had some custom training done to it, and while it still works well, it was created more for my personal work than for public release. It leans more towards studio backdrops and my custom training.

v7 - Similar faces and bodies to v6, but a little more flexible. A good in-between of 5 and 6.

About RyanBlend:

The RyanBlend models are designed primarily to create photorealistic images of women with a high-end glamour/boudoir photography style, with a very subtle hyper-realistic anime style for facial features and body proportions. It's NSFW friendly, but doesn't have to be used that way.

It's excellent at creating closeup skin texture and high quality faces without using a face restoration model.

I've been a professional photographer since the mid-2000's and specialize in a high-end retouched look for glamour and erotic photography. So I try to make my RyanBlend models close to my own photography and editing style.

The base of the merge consists of:

Deliberate, phDMix, and Realistic Vision v2

...plus a little:

Analog Madness, HARDblend, URPM, and Perfect World

v6 & v7 also have a lot of my own custom appearances for faces and bodies.

I recommend using VAE ft mse 840000

A merged model trying to get as close to realistic as I can.

V1 contained parts of Deliberate, Uber Realistic Porn Merge and Realistic Vision

V2 used updated Deliberate, Realistic Vision and Uber Realistic Porn Merge.

V3 added RPG, HassanBlend, Dreamlike Diffusion, Babes (experiment 8), Experience (realistic), Protogen x5.3 and Protogen x5.8 to the original 3 models.

V3.2 used updated Realistic Vision, Experience and switched to Babes Experiment 9.

V4 has been redone from the ground up. The model will consist of Deliberate, Realistic Vision, Experience, RPG, Analog Madness and CyberRealistic. This will come in 3 flavours, Standard, LE edition and soon a vivid edition for those who like the bolder colours seen in V3.2.

Recommend: DPM++2M Karras, Steps: 32+, Hires fix 0.3 - 0.5, CFG 6 - 8.

I would also recommend using epi_noiseoffset https://civitai.com/models/13941?modelVersionId=16576

Check out more of my work over at Deviantart - https://www.deviantart.com/consistentfactor444

Posts of images welcome.

Feel free to leave a review about the model.

If you would like to see more consider supporting me in the linke above.

Used as base models:

Clarity:V2: https://civitai.com/models/5062/clarity

ConsistentFactor:V4.0: https://civitai.com/models/9114/consistent-factor

Colorful:V2.6: https://civitai.com/models/7279/colorful

Do you want to use artists?

https://stablediffusion.fr/artists

• Adding [by Artist, etc] at the end of the prompt can give you the desired result

Nicknamed: C3

Credit to justmaier

Version info can be found under the review section.

I hope you enjoy.

This is a mix model based on:

• https://civitai.com/models/47067/pikas-new-generation-v10

• https://civitai.com/models/4468/counterfeit-v25

The settings I recommend：

Res:512*768

Sampler:DDIM

Step:25

CFG Scale:7

Clip Skip:2

Hiresfix:ON and set denoise to 0.4-0.6

VAE:any anime

A simple Mix,nothing fancy,just plain and simple.

Hope you enjoy🧐

Sweet-mix

Sweet-mix is the spritual succesor to my older model Colorful-Plus. (on hugging face)

What sweet-mix performs in comparison to my older model is probably everything, seriously, check it out, i think it's really neat.

Models that went into this merge:

• Anything-v4.5

• AbyssOrangeMix 3

• Protogen

• CitrineDreamMix (probably)

• Peachmixs

• Sardonyx-Blend (introduced in v1.1)

• Meinahentai (introduced in v1.2)

• IvoryOrangemix (introduced in v1.3)

• MixPROV4 (introduced in v1.4)

• Daintymixhentaimodel (introduced in v1.4)

• Jasminiquemix (introduced in v1.5)

• Sirenmix (introduced in v1.6)

• bbmixluci (introduced in v1.6)

• Sardonyx-Blend Redux (introduced in v1.6)

• IceRealistic (introduced in v1.6)

MUST-HAVES FOR GREAT IMAGE GENERATION LIKE IN THE IMAGES:

This negative prompt: (bad_prompt_version2:0.8), EasyNegative, (worst quality:1.3), (low quality:1.3), extra fingers, fewer fingers, bad-hands-5, bad-artist,

These textual inversions:

- bad-artist (Not the anime version)

- bad-hands5

- bad_prompt

- easy_negative

Blessed2-VAE (Thanks Nocrypt, that's a really good VAE)

Clip skip: 2

And a great imagination :)

Mix between chilloutmix and perfectWorld with a bit of koreandolllikeness

Her name is RUIS.

==ReadMe===========================================

The fun AI life we all enjoy

let's have fun

==================================================

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.

A sample of commonly drawn images

A mixed model I made from taking the first models I could merge and passing them down constantly until I got to this one.

This is more of a personal model but I felt like sharing it with everyone since it produces pictures and does things other models weren't doing for me.
It's mostly focused on women and furry stuff, trying to get a good mix between all the worlds while always defaulting to an artistic style.

Words of warning when using it, without using the proper prompting style, it will give some middling to bad results.
To use the proper prompting setup: [Umbrella terms] [Type, Subject] [Tags, Description of the scene] [Artist/Art style]

The method of mixing probably doesn't matter but I mixed the models down until they looked horrible then mixed them back up till they gave differing results that looked good.

VAE wise I'd suggest using an anime one.

" IceJi ProFantasy was tested with a huge of scenarios, and fine-tuning several times to ensure that you will be able to create a FANTASY image in your own way "

This model merges with my own 3 models called R-base (realistic), F-base (Fantasy), and V-base (Vibrant) mixed together with 3 formulas Real, Mix, and Max (but in v2, these will combine into one)

** I suggest using the FULL version for a better experience **

Overview:

• The purpose of this model is to generate epic fantasy images not only Realistic but Animation also.

• These models require a VAE (You can try mse-840000-ema or anything you want)

Recommends:

• You use it with any steps, prompt, sampler, ... (anything you want)

• Steps: I recommend 30-40, 60-80, and 120-150 (around 100, I think it's not well)

• Sampler: I developed it with DPM++ 2S a Karras, so I recommend this sampler (but others also work fine)

• Sizes: 512x512, 512x768, 512x904 (Can be used both vertically and horizontally). Not recommended sizes more than 768x768

• Hires. fix: Use Denoising strength less than 0.6 (but you can try >= 0.7, sometimes you will get problems with the hands)

• Lora: It works perfectly with VFX Lora by @DitamAi, and EPI Noise Offset.

  • I do not recommend using Face Lora because you will lose a fantasy style but you can try if you know what you want from it

Versioning:

• v1: pilot version

  • v1-Real: Cinematic style, Realistic photo, less fantasy (if you have a fast GPU and you will generate a batch of many images, I suggest this version)

  • v1-Max: Ultra Fantasy, Concept Art, Animation, Semi-realistic (If you want semi-realistic or you plan to use img2img to make the image look more realistic, I suggest this version)

  • v1-Mix: Mix both versions (This is for lazy people who want something in the middle)

• v2: All of the above, can generate everything (depending on your prompt)

• v3 (planned): Have only base models and use 512-Network Lora (or hyper network) to create a unique and specific image style type.

Comparison:

# Version 2: Full vs Pruned

full-size image

# Old versions

full-size image

# Steps

full-size image

# Samplers (with steps 45)

full-size image

Disclaimers:

• Do not sell: on any website.

• Credit: If you use my model in your own merges

• Not authorized: to be used on generative services (without my permission)

中文/English

这是一个混合模型，如果有需要改进和做得不对的地方请指出来（如果你能说出纠正的方法最好，因为我可能不是很清楚）。

这个模型的方向是尽可能忠于提示词,能生成更多的东西(人物是主要的,但不代表其他事物的效果不重要),舒服的光影。

从AOM3:https://huggingface.co/WarriorMama777/OrangeMixs 和Anything:https://civitai.com/models/9409/or-anything-v5 知道了很多模型和学习了很多混合方面的经验,十分感谢。

使用建议:

如果在脸部不是焦点的画面中(非正面也是同样),脸部出现崩坏的情况,以及想提升面部质量,可以使用Hires. fix获得更好的面部,使用其他随机种或者工具也是不错的办法。(我认为这是混合了3D或者2.5D模型导致的,并没有什么很好的办法解决,如果你有办法在模型层面改善请告诉我。)

较高的分辨率(比512 * 512高一点)再加上Hires. fix,图片质量会更好(如果显存不够你可以尝试低倍率的Hires. fix或者其他放大方法)。

过高的,用于画面质量(像best quality这些)的正面和负面提示词是不必要的,会减少细节,也可能使画面趋于一种风格,以及无法对提示词做出反应。

如果觉得画面内容不够丰富,你可以尝试细致地描述,使画面更加贴近你的想象。

如果想要更强的明暗对比效果,可以尝试光影类lora,比如epi_noiseoffset:https://civitai.com/models/13941/epinoiseoffset (我曾在V1加入过它,但是它似乎会影响手指生成,所以去掉了,你可以自己使用它)

如果你想要更多的亚洲面孔,人物类lora是很好的选择,比如Doll Likeness系列:https://huggingface.co/Kanbara/doll-likeness-series

This is a merge model, so please point out if there is something that needs to be improved and done wrong (If you can say how to correct the best, because I may not be very clear).

The direction of this model is to be as faithful as possible to the prompt words, to be able to generate more things (the character is the main, but does not mean that the effect of other things is not important), comfortable light and shadow.

I know a lot of models and learn a lot of mixing experience from AOM3:https://huggingface.co/WarriorMama777/OrangeMixs and Anything:https://civitai.com/models/9409/or-anything-v5, thank you very much

Suggestions for use:

If the face is not in focus in the Image (same for non-frontal), the face appears to be falling apart, and you want to improve the quality of the face,you can use Hires. fix to get a better face, or use other seed or tools is also a good way.(I think this is because of the merge of 3D or 2.5D models,There is no good way to fix it,If you have a way to improve it at the model level please let me know.)

Higher resolution (a little higher than 512 * 512) plus Hires. fix, picture quality will be better (if the gpu memory is not enough you can try a Low magnification of Hires. fix or other upscale tools)

Overly high, positive and negative prompt for image quality (like best quality) are unnecessary and will reduce detail, and may tend to make the image look a specific style and the inability to respond to the prompt.

If you feel that the content of the picture is not rich enough, You can try to describe in detail to make the picture more closely to your imagination.

If you want a stronger contrast between light and dark, you can try light and shadow lora, such as epi_noiseoffset:https://civitai.com/models/13941/epinoiseoffset (I had added it in V1, but it seems to affect the finger generation, so remove it, you can use it yourself)

If you want more Asian faces, the Character class lora is a good choice, such as the Doll Likeness series: https://huggingface.co/Kanbara/doll-likeness-series

你可以从版本信息中知道当前版本信息,比如使用了哪些模型/You can know the current version information from the version information, such as which models are used

我使用这个VAE/I use this VAE:https://huggingface.co/stabilityai/sd-vae-ft-mse-original

配方在/Recipe in → huggingface:https://huggingface.co/Fre2C/DreaMirror-Mix

AngrA nim version to try generate less realistic

(Angra_Anim.Flex.v01)

-My 2nd Model -Checkpoint Type merge

If you feel like supporting my work, please consider visiting my BuyMeaCoffe

Trying add more hyper realistic in the same recipie to make flexibility to create a still cool image, to achieve beautiful realistic scenes.

It works also to making general or NSFW images of models of high quality.

Recommend: DPM++2M Karras, Steps: 20+, Hires fix 0- 0.5, CFG 6 - 9.

scottie style model:

v1.0: preliminary test version

v2.0: complete test version

v3.0&3.5: improved version based on previous iterations

v4.0: official series

v5.0: test experimental version based on v4.0

v6.0: fully optimized version

i am continuously improving this model and welcome you to download it.

NeverEnding Dream (a.k.a. NED)

"This is a dream that you will never want to wake up from. A NeverEnding Dream."

Add a ❤️ to receive future updates. This took much time and effort, please be supportive 🫂
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.

It has been a while since I made a model. I've been making LoRA's non stop for the last month, but I felt like I needed something. A big problem with DreamShaper is that it wasn't much compatibles with the LoRA's I was making. So I wanted to create a model that was "artistic" like DreamShaper but also able to use my LoRA's and booru tags for generating images. Plus I wanted this to be able to make good anime stuff out of the box, without needing an additional LoRA.

After a lot of tests and also fine tuning (plus LoRA merges) here is my newest model.

It came out a bit more realistic then I wanted, but I won't complain.

What is this model great at?

• Generating cosplay images

• Generating anime pictures

• Work accurately with character LoRA

• Generating good looking people

• Generating realistic animals

• Generating images using booru-like tags

What is this model not too good at?

• Generating pictures from complex sentences

• Making fantasy/sci-fi paintings

So this basically complements DreamShaper, and it also doesn't include Dreamlike in the mix.

I hope you'll enjoy it!

Some tips:

• Use CLIP skip 2.

• Select "auto" as vae if you're using the baked vae version.

• Use highres fix or img2img to upscale the images after you get a preview. See the examples generation data for settings suggestions, I've tested various techniques.

• If you're making images where the subject is far, remember you can inpaint eyes and faces selecting "only masked".

• Stable Diffusion is great, have fun with it!

IF YOU ARE NEW TO STABLE DIFFUSION OR YOU HAVE PROBLEMS WITH MY MODELS, GO HERE: Guide

What's new in V2?

Emerald is my least downloaded checkpoint and personal least favorite. It was too focused on a single style, and all the characters were really similar to each other. For V2 I've rebuilt the dataset from scratch using a custom LoRA trained on the pictures I liked from the previous iteration and adding an unreleased LoRA to make it more vibrant. The result is a more flexible model with less over saturated colors and easier to use with higher CFG values.

How to use this

If you are new to the world of Stable Diffusion, you can learn how to run it by yourself here:

Tutorial

You'll also need a VAE to use this. For the examples, I've used the Grapefruit one:

https://huggingface.co/iZELX1/Grapefruit/resolve/main/Grapefruit.vae.pt

What's this model

Mistoon_Emerald is a variant of my SD model Mistoon_Anime that tries to achieve a more "fan art cartoon" style while still maintaining some resemblance to the modern anime style.

The starting point for this model was my blend. From there, I then added a few of my custom LoRAs in the mix to change the esthetic.

Try also the other flavors:

The style

It creates realistic and expressive characters with a "cartoony" twist. Unlike other anime models that tend to have muted or dark colors, Mistoon_Emerald uses bright and vibrant colors to make the characters stand out.

All the examples have been created using this version of the model using my workflow. Check the guide if you want to learn more about my workflow.

The model can also easily do NSFW stuff (check the last examples).

I highly suggest using a low CFG (3-5) value to avoid the usual burn effect.

Guides

Learn how to create pictures like mine with my step-by-step tutorials:

• Beginner's Guide

• Prompting

• My Workflow

• Inpainting

Support

If you want to support my work, you can check out my music here:

https://inzaniak.bandcamp.com/

You can also check my AI Art accounts, where you'll find daily pictures made with my models (and previews of new ones coming):

• Instagram

• DeviantArt

Identity Disorder

"IDENTITY DISORDER" - as IN Dissociative Identity Disorder.

If you want to learn more please go here: https://thepluralassociation.org/ and support us, because we're being fake claimed into oblivion for "not being ashamed". Never be ashamed if you have quirks.

No, the model literally doesn't have it but we do, and we wanted to MEME yet be a little artsy serious about it in prompting.

If you got requests, or concerns, We're still looking for beta testers: JOIN THE DISCORD AND DEMAND THINGS OF US: https://discord.gg/Da7s8d3KJ7

Listen to the music that we've made that goes with our art:
https://open.spotify.com/playlist/00R8x00YktB4u541imdSSf?si=b60d209385a74b38

We stream a lot of our testing on twitch: https://www.twitch.tv/duskfallcrew

any chance you can spare a coffee or three? https://ko-fi.com/DUSKFALLcrew

What's in the BOX:

• Boneless Unreality Mix

• Maple Syrup @advokat

• The model that i can't pronounce but has Samdoes Arts AND Inkpunk @xerminator13

• Something 2.2 which i think is a mix by @nocrypt

• All the berry mixes I could find

• Dark Sushi 2.5d

• Hentai, Elysium, NAI/Anything 3

• Osenayan 1.5 Beta Illustration

• Whatever's in the above model and it's loras XD

This is a different take on the 2.5d anime, and will be built upon as a series in future.

As always smack the ko-fi button if you want us to stop making models - I MEAN AHEM DO IT BECAUSE YOU CARE AND WANT US TO BE ABLE TO LIVE AGAIN (I think my models are taking me hostage)

PERSONAL USE A OK - GENERATION AND COMMERCIAL SITES PLEASE ASK - DO NOT RESELL THIS MODEL.

Downstream merges: Idgaf what you do, just take a stand about models getting uplifted without consent is all I care!

Introduction

This model is used to generate realistic architecture rendering works, with a focus on human perspective and a balance between bird's-eye and indoor performance.

The prompt word can be in the form of tag or natural language, but it is not recommended to input too long prompt words. Welcome to share your own works (with 5 ☆).

I don't know why my pictures uploaded are always missing generating data, you can down load it for the full generating information.

Recommended Parameters:

prompt:modern architecture style, photo realistic, hyper detailed photo, human perspective,

negative prompt:lowres, normal quality, worst quality, cropped, blurry, drawing, painting,

说明

本模型用于生成各类建筑写实渲染效果，以人视效果为主，兼顾鸟瞰与室内的表现。

提示词以tag形式或自然语言均可，但不建议输入太长的提示词。欢迎分享作品自己的作品☆，

Better with EasyNegative embedding network: https://huggingface.co/datasets/gsdf/EasyNegative

Attention: You need to get your own VAE to use this model to the fullest. While it does work without a VAE, it works much better with one. I recommend you try this one out.

Hey everyone. After GrapeLikeDreamFruit hit, I started missing having a more general purpose model for the mundane kind of pictures - nude photographs on different backgrounds and some light hardcore capabilities. This model here is my response to that need. It handles the female nude superbly, and while it's less of an artistic model than GrapeLike, it's still quite capable in that regard. It's quite good at hardcore, even if that is just a secondary goal for this model, and can be prompted for a variety of acts. Model has a good response to Danbooru tags.

This model involves dreamlike photoreal, so here is the license that you must abide by.

See examples for some ideas.

Deeper explanation:

This model merge was done following a similar phylosophy to GrapeLike: a Block merge between a realistic anatomy + skin core and a posing anime core. The block merge is done in such a way as to emphasize the anime core in the center of the Unet, while rapidly decaying back to the realistic core at the edges. This has the effect of bringing a lot of posing and composition ideas from hentai models inside the photorealistic core we have available without touching textures and photorealism. Full recipe follows:

Anatomy core:

izumi, F222, dreamlike photoreal, realistic vision 1.2, sxd, all at the same intensity. ie, the merge chain was:
(((izumi + F222 0.5) + dreamlike photoreal 0.33) + realistic vision 0.25) + sxd 0.2

Anime core:

Anything v4.5 merged with Basil Mix using the same block merge coefficients used in mixing Abyss Orange Mix, then merged 40% with grapefruit, the result was merged 30% with gape60 and finally 15% with RPG v4.

Bringing both together:

The block merge for both was done using a formula. I kept the bottleneck Model A was anatomy, model B was anime. I kept the center lalyer at 0.7, as well as base alpha, then followed 0.8/(n**1.1)as a merge rule, with n being distance from the center. Full numbers were "0.05199847612695355,0.05722134125067434,0.06354625877794251,0.07135480548979826,0.08122523963562354,0.09407671474206218,0.11146117361039158,0.13621438760332552,0.17411011265922482,0.23892225595753655,0.373213196614723,0.8,0.7,0.8,0.373213196614723,0.23892225595753655,0.17411011265922482,0.13621438760332552,0.11146117361039158,0.09407671474206218,0.08122523963562354,0.07135480548979826,0.06354625877794251,0.05722134125067434,0.05199847612695355".

Base on linneamix and nijiv5style

Cute, good,ecchiable

U know, I can't say anything about Ⅼ○ⅬⅠ

可以随机的创造充满东方异域神秘感的符号，用符号进一步引导出图，可以形成纹理丰富的作品！

You can randomly create symbols full of oriental exotic mystery, and use symbols to further guide the picture, and you can form works with rich textures!

Miaoshou Civitai Assistant | 喵手助理

MiaoshouAI Assistant for [Automatic1111 WebUI]

喵手助理 [Automatic1111 WebUI]

Features | 功能:

1. Download Civitai models from your webui, miaoshou assistant will automaticly set the model cover and put the models in the right folder for you.

   喵手助理非常簡單的搜索並下載Civitai或者煉丹閣上面的模型，同時會自動保存模型的封面和將模型自動下載到對應的文件夾中。

1. Manage your existing model, get your trigger word and model sample image tags easily. Send them to your prompt to start edit your own.

   喵手助理可以管理你的模型，讓你查看模型的觸發詞，並且一鍵將模型的tag信息發送至txt2img, img2img, inpainting, extra 的提示詞。

1. Download different Stable Diffusion Official 1.5, 2.1 Models, Controlnet Models and VAE models from one place.

   你也可以在模型源中，選擇下載不同的官方模型的1.5 2.1模型文件，Contrlnet以及VAE的各類模型。

1. Manage your boot setups including CPU settings, theme settings, API and port setting, etc with simple clicks.

   啟動助手可以幫助你輕鬆一鍵設置並保存CPU，主題，API和端口等的啟動設置。

Installation | 安裝:

1. Install from Automatic1111 WebUI | 在Automatic1111 WebUI 下安裝

In Automatic1111 WebUI, go to Extensions Tab->Install from URL, copy the following address in "**URL for extension's git repository**".

在 Automatic1111 WebUI 中，前往 扩展插件-> 从URL安装，在`扩展插件的git仓库网址`中复制以下地址。

https://github.com/miaoshouai/miaoshouai-assistant.git

Click Install, wait until it's done. Go to Settings-> Reload UI

点击安裝，等待安装完成。然后前往設置-> 重新加载界面

1. OR you can just click on download from this page and unzip your file in your extensions folder under webui folder, then restart your Webui.

   或者你可以直接點擊下載本頁面上的zip文件並解壓至你Webui目錄下的extensions文件夾

I've made some significant updates to the LusciousMix 1.5 version of the model in order to add more diversity and improve the overall quality of the generated photos. Even though I've introduced more variety, rest assured that this version is still capable of generating stylized or semi-realistic photos like the previous version.

I want to note that I am not providing a smaller 2gb pruned-fp16 version this time, as the quality prune version did not generate good-quality images. I don't want to upload a version that doesn't meet my standards, so I'm not releasing that version. Additionally, I've taken feedback from the community and made significant improvements to the 1.5 version, including better image resolution that supports more prompts and generates more realistic photos than the previous version.

I'm excited to share that this version also excels at Euler A sampler and DPM++ 2. However, please keep in mind that this is still an experimental version, and I'm working hard to continue improving it. If you don't like the full version, please do not download this model.

Your feedback is valuable to me, so please don't hesitate to share your thoughts on this version so I can improve future iterations. Thank you!

The model merge has many costs besides electricity. With your support, we can continue to develop them.

モデルマージは、電気代以外にも多くのコストがかかります。皆さんの応援があれば、これからも開発を続けることができます。

https://ko-fi.com/breakd

I have merged this model with a lot of background and detail.

The output of this BreakDomain series will be more like an illustration or animation than BreakDro.

背景や細部の緻密な描写にこだわってマージをしています。

このモデルシリーズではBreakDroよりもイラストやアニメ寄りの出力が期待できます。

No baked VAE.
The recommended VAE is "vae-ft-mse-840000-ema-pruned.ckpt".
If you try it and make a good one, I would be happy to have it uploaded here!
The merging source is listed in each version detail.

VAEは焼いてません。
おすすめのVAEは"vae-ft-mse-840000-ema-pruned.ckpt"です。
使ってみて、いいのができたら、ぜひここにアップしてもらえると嬉しいです！

I have gone back to the drawing board and redone the merge to remove the Creator Credit Required license from BreakDro. this is a new model that is finished to the same level as BreakDro.

BreakDroからCreator Credit Requiredライセンスを取り除くために、初心に戻ってマージをやりなおしました。BreakDroと同等に仕上げた新作モデルです。

Future updates will be made to this model.

今後の更新はこのモデルを行っていきます。

Brainime

Will produce almost anything (as long as it's anime) without too much of a prompt, see Notes below.

This model prefers natural language than long comma-separated chunks. It will still work with it, just not as well!

Your feedback is essential to improve the future updates, please consider leaving a comment or a review 🙏

Recommended Settings:

• Sampler

  • DDIM

  • DPM++ variants

• CFG Scale

  • 3~7 (I leave mine around 5 usually)

• Steps

  • With Hi-Res: 20~30

  • Without Hi-Res: >=50

• Hi-Res highly recommended

Recipe

This is my first merge, I've sort of lost the recipe at this point but it's using mostly

• CoffeeMix

• BluePencil

Notes

• Highly recommended to use EasyNegative

I'm still new to this so I'll add more details on here as I get feedback from the community.

What's new?

-Better Hands than the other Crescent Models

-Better Color than last Crescent Models (CrescentHORIZON)

-High quality ouput

Check my gallery on :

https://id.pinterest.com/vextoria_studio/

NEGATIVE PROMPT RECOMMENDATION :

(3d, render, low quality, worst quality, bad quality, ugly, low resolution, nudity, nsfw, paintings, anime, cartoon, drawings:1.3)

INFORMATIONS

The other upcoming Crescent Project will be released with a name instead of the number of the version.

3.CrescentHORIZON (Release)

Focus on hyperrealism, merged with EpiNoiseOffsetPY. so the ouput probably more darker than the previous Crescent

---------------------------------------------

4.CrescentORBIT (NOT RELEASED) *too much bug

5.CrescentGAUNTLET (Released)

In case someone wishes to use the model I'm using here, it can be obtained here. This model is specifically designed for realistic photography and similar applications. It supports the Indonesian language, but it may require some English supplementation.

Just in case ada yang pengen nyobain model yang aku pake, aku upload disini. Model ini khusus dirancang untuk fotografi realistis. Model ini juga mendukung bahasa Indonesia, namun masih perlu didampingi dengan bahasa Inggris.

Goal of this project

• Realistic and Detailed models

• General use for photograhy

• Support Indonesian Languages

Recommended setting for this model

VAE IS RECOMMENDED : https://huggingface.co/stabilityai/sd-vae-ft-mse-original/tree/main

Sampler : Euler A (HI - RES FIX REQUIRED) / DPM++ 2M KARRAS

Steps : 30 Steps / 15 is ok but not recommended

CLIP SKIP : 1 / 2

Hi-res fix (very recommended)

**You can't use the model to deliberately produce nor share illegal or harmful outputs or content**

Semoga bermanfaat, Terima kasih telah mencoba & menggunakan model ini <3

Please FEEL FREE to comments about this model below, or if you have some recommendation that would be really appreciated for this project!

Feel free to mix or blend this model with your models, if you upload it, please let me know. I wanted to try yours too <3

Contact :

www.instagram.com/promptworld

Hide

Just another merge of some popular models, that I think turned out good.

It's somewhere between 2.5d and realistic. The merge is a bit biased towards nudity, blonde hair and Asian looks, but is still pretty versatile. (see example images)

There is no VAE baked in, so I suggest using either vae-ft-mse-840000-ema-pruned.vae.ckpt or kl-f8-anime2.ckpt with the merge.

The merge works well with either clip skip 1 or 2.

Used models:

uberRealisticPornMerge_urpmv13

henmixReal_v23

cyberrealistic_v20

revAnimated_v11

deliberate_v2

Edit:

Added a pruned version.

Its a merge of PerfectWorld and Deliberate with better faces, hands & eyes

欢迎大家来评论

在copper模型的基础上使用了mbw插件继续融合

这个模型很容易画出看起来娇小的角色

使用推荐

正面tag建议不要使用medium breast，small breast的效果更好

负面tag推荐使用Easynegative（一个负面pt模型）：https://huggingface.co/embed/EasyNegative/tree/main

例如：nsfw,EasyNegative,nsfw,(low quality, worst quality:1.4), (bad anatomy), (inaccurate limb:1.2),bad composition, inaccurate eyes, extra digit,fewer digits,(extra arms:1.2),2girl,lanjiao5,text

VAE推荐挂载通用VAE(即Animate.vae)，使用vae-ft-mse-840000-ema-pruned这种写实风格的vae也可以，不过偶尔会有颜色过重的情况。

v1.5

相比于1.0变化不大，解决了一些颜色过重的问题，1.5版本的颜色会更浅一些

如果觉得颜色不够鲜艳推荐使用clearvae:ClearVAE | Stable Diffusion Checkpoint | Civitai

同时也改善了一些nsfw场景，但不多（

v2.0

在尽可能保留模型特色的基础上，改善了一些人物细节，增强了背景效果

v3.0

与2.0相比有了一定的变化，但不多，主要是丰富背景，如果你需要看起来更幼态，还是推荐使用2.0或者2.5

Suggested negative prompts for photorealistic: (semi-realistic anime illustration 3d rendering drawing)

Beauty in the ordinary.

There are tons of checkpoints to achieve a "pretty" or "model-like" face.

It's time for a checkpoint that makes the ordinary and common, like a snapshot taken in passing, feel more real.

That's why ANINDE-mix was born.

It's not just a checkpoint to draw pretty people,

but about creating naturalistic realism with fewer prompts.

AND also very friendly for NSFW content.

With v.GLO, you get realistic Western people,

v.ASI to create realistic (not so ugly) Asians.

GeminiX_Mix is a high quality checkpoint model for Stable-diffusion, made by Gemini X.

Warning: This model is NSFW.

This is the first model I have published, and previous models were only produced for internal team and partner commercial use. Now I am sharing it publicly.

This module supports the generation of characters of various races from around the world, including Caucasians, African Americans, Latinx, and Asians.

You can use it for any purpose, but it cannot be used for the following:

Training to generate celebrity headshots.

Producing products that exploit minors.

Supporting terrorism-related products.

Producing violent or gory products.

--------------------------------------------------------------------

Below are the parameter settings that I recommend:

Prompt:

(masterpiece, extremely detailed CG unity 8k wallpaper, best quality, Amazing, finely detail, cinematic lighting, highres, ray tracing), {an extremely delicate and beautiful, RAW, perfect anatomy, best shadow},

1girl, solo, black hair, braid, (freckles:0), (large breasts:1.0), naughty expression, pale skin, black dress, pearl necklace, Pearl ear-rings, parted lips, kpop, {{floating hair}}, dynamic pose,

cityscape, outdoors, detailed background, Cyberpunk world, Modern metropolis, neon, Sci-fi style,

Negative Prompts:

(worst quality, normal quality, low quality:1.7), watermark, signature, text, freckles, lowres, monochrome, grayscale, bad proportions, ugly, dirt on face,

Size:512X768

Steps: 30

Sampler: DPM++ SDE Karras

CFG scale: 13

Hires.Fix(check)

Lanczos

Hires steps : 0

Denoising strength : 0.30

Upscale by : 1.5

--------------------------------------------------------------------

Dozens of AICG creators have already used this model with positive feedback. If you have any questions, please feel free to follow my Twitter and discuss with me:

https://twitter.com/CuteBlueEyed

******************************************************************************

GeminiX_Mix 一個高品質的Stable-diffusion checkpoint模型, 由Gemini X產出.

(這是一個NSFW模型)

這是我發表的第一個模型，以前所產出的模型僅作為內部團隊及合作夥伴商業使用，現在我將它分享出來。

這個模組支援世界各人種人物的生成，包含歐美白人，黑人，拉丁種族及亞洲人。

你可以使用他做任何事，但是不可以涉及下面事項:

1.不可涉及名人頭像訓練產出。

2.不可涉及未成年兒童性剝削產品產出。

3.不可涉及恐怖主義產出支援。

4.不可涉及血腥暴力產品產出。

--------------------------------------------------------------------

下面是我推薦的參數設定:

Prompt:

(masterpiece, extremely detailed CG unity 8k wallpaper, best quality, Amazing, finely detail, cinematic lighting, highres, ray tracing), {an extremely delicate and beautiful, RAW, perfect anatomy, best shadow},

1girl, solo, black hair, braid, (freckles:0), (large breasts:1.0), naughty expression, pale skin, black dress, pearl necklace, Pearl ear-rings, parted lips, kpop, {{floating hair}}, dynamic pose,

cityscape, outdoors, detailed background, Cyberpunk world, Modern metropolis, neon, Sci-fi style,

Negative Prompts:

(worst quality, normal quality, low quality:1.7), watermark, signature, text, freckles, lowres, monochrome, grayscale, bad proportions, ugly, dirt on face,

Steps: 30

Sampler: DPM++ SDE Karras

CFG scale: 13

Hires.Fix(check)

Lanczos

Hires steps : 0

Denoising strength : 0.30

Upscale : 1.5

--------------------------------------------------------------------

已經有數十位AICG創作者使用這個模型，大家的反應都不錯，有任何問題歡迎追蹤我的推特與我討論:

https://twitter.com/CuteBlueEyed

Trained model using 50+ images of Elle Fanning

it is highly recommended to use it in conjunction with this embedding.

https://huggingface.co/datasets/stb/embeddings/blob/main/3113fann1ng3.bin

Check prompts.

铭刻，漫步于樱花丛中的时光。这里是来自于galgame《樱之刻》的原画级大模型。

Time of walking among the cherry blossoms. Here is the original painting-level model from galgame Sakuranotoki.

特点是干净又卫生通透，眼眸与头发丝的高光非常细腻。

This model features clean, transparent, very accurate highlights on the eyes and hair.

使用时，以下配置可能比较有用（其他的用法待开发，欢迎尝试和分享）：

When using it, the following configuration may be more useful (other usage methods are to be developed, welcome to try and share):

DPM++ 2M Karras/DPM++ SDE Karras

Hires.fix/BSRGAN

CFG scale=6~9

这个融合模型可以生成多种多样风格的插画，整体风格偏向动漫写实渲染，对于面部，会根据你的词条改变风格（我个人比较喜欢可爱的风格，所以往里面融了一些大家喜闻乐见的脸模），这个模型对各种词条有还算不错的兼容性，如果你有什么好玩的想法，可以拿它试一试，希望你能享受这个模型带来的乐趣~

我常用的配置是:Steps: 24, Sampler: DPM++ 2M Karras, CFG scale: 7, Size: 512x768, Denoising strength: 0.3, Clip skip: 2, ENSD: 31337, Hires upscale: 1.5, Hires steps: 10, Hires upscaler: R-ESRGAN 4x+ Anime6B

以下内容为机器翻译/The following is machine translated:

This fusion model can generate a variety of styles of illustration, the overall style is inclined to anime realistic rendering, for the face, will change the style according to your words (I personally prefer a cute style, so into the fusion of some of the popular face models), this model has a pretty good compatibility with a variety of words, if you have any fun ideas, you can take it to try, I hope you can enjoy the fun brought by this model ~

My usual configuration:Steps: 24, Sampler: DPM++ 2M Karras, CFG scale: 7, Size: 512x768, Denoising strength: 0.3, Clip skip: 2, ENSD: 31337, Hires upscale: 1.5, Hires steps: 10, Hires upscaler: R-ESRGAN 4x+ Anime6B

The model nobody wants! A well trained Dreambooth (JP repo) model of the MyPillow CEO Mike Lindell. For best results use Hi-Res fix. For ultimate results combine w/ Controlnet.

Renders fantasy art in a style similar to Larry Elmore / Dragonlance, both groups and portraits

I remade Sardonyx. It's about damn time.

This is an NSFW prone model. It is meant to generate primarily women. Instead of using AOM3 as the base model, I merged together my own base model with more modern contributions. Additionally, I manually re-made all the style LoRA's I drew from to make this model. I did a lot to ensure quality.

I use Hires Fix with 4x Fatal Anime 50000 G at 15 Hires steps and 0.45 Denoising strength, upscaled by 2 times. I use the Orangemix VAE. Everything else is in the tags.

I also have a Discord. Join if you wanna share results, get advice, or make requests.

IF YOU ARE NEW TO STABLE DIFFUSION OR YOU HAVE PROBLEMS WITH MY MODELS, GO HERE: Guide

How to use this

If you are new to the world of Stable Diffusion, you can learn how to run it by yourself here:

Tutorial

You'll also need a VAE to use this. For the examples, I've used the Grapefruit one:

https://huggingface.co/iZELX1/Grapefruit/resolve/main/Grapefruit.vae.pt

What's new in V2

V2 is a complete remake of the original Ruby checkpoint. Starting from Mistoon_Anime I've re-balanced the mix of LoRAs to reach a more flexible result. The new version can easily work with higher CFG levels than the previous one, and is also closer to the style I had in mind when I made this checkpoint.

What's this model

Mistoon_Ruby is a variant of my SD model Mistoon_Anime that tries to achieve a more "western cartoon" style while still maintaining some resemblance to the modern anime style.

The starting point for this model was my blend. From there, I then added a few of my custom LoRAs in the mix to change the esthetic.

Try also the other flavors:

The style

It creates realistic and expressive characters with a "cartoony" twist. Unlike other anime models that tend to have muted or dark colors, Mistoon_Ruby uses bright and vibrant colors to make the characters stand out.

Mistoon_Ruby is ideal for anyone who loves western cartoons and animes, and wants to blend the best of both worlds.

All the examples have been created using this version of the model using the HiRes-Fix. Check the guide if you want to learn more about my workflow.

The model can also easily do NSFW stuff (check the last examples).

I highly suggest using a low CFG value to avoid the usual burn effect.

Guides

Learn how to create pictures like mine with my step-by-step tutorials:

• Beginner's Guide

• Prompting

• My Workflow

• Inpainting

Support

If you want to support my work, you can check out my music here:

https://inzaniak.bandcamp.com/

You can also check my AI Art accounts, where you'll find daily pictures made with my models (and previews of new ones coming):

• Instagram

• DeviantArt

Changelog

2023/03/31: Added Inpainting Model

Realistic Photo and Anime

It is a inpainting merge of the Char-V10 model, which was trained on some 1024px images.

Realistic+Anime+Draw3D

+https://civitai.com/models/52297/rmega-anime-v7

+https://civitai.com/models/45569/v15amixphotofp16

A sister project of WyvernMix, geared more towards NSFW erotic portraits.

Works well with 5-7 CFG, prefers higher native res (768x960)

Doesn't do too well with more hardcore stuff, but maybe it just needs better prompting, let me know if you have good results.

Also let me know in the comments what improvements I could make in future versions.

Mix Realistic photo+Anime

Thinner pencil, many thinner and softer!

THIS MODEL IS A SPECIAL MERGE AND TRAIN MIX OF MY OWN TRAINED LORA + THE CRAZY EXPRESSION LORA + OTHER MERGES!

you can use the keywords for crazy expression lora with this model and with the right offsets it will work!

• 适当调小(CFG Scale)的数值，可以得到与众不同的效果。

V2版本说明:

1.更换了默认脸。

2.更好的光影表现。

3.融合了更多模型，后面还会继续迭代。

4.如果想得到更好的皮肤质感，那么生成图片的时候不要开启高清修复，而是将生成的图片传入图生图模式，启用Tiled Diffusion，来进行放大，从而得到更真实的皮肤质感。

5.我会继续更新迭代这个大模型，希望能不断填充内容，让模型更具趣味性。

谢谢您的支持！

Version 2 Release Notes:

1. Changed the default face.

2. Improved lighting and shadow effects.

3. Added more models and will continue to iterate in the future.

4. For better skin texture, do not enable Hires Fix when generating images. Instead, use the "Tiled Diffusion" mode to enlarge the generated image and achieve a more realistic skin texture.

5. I will continue to update and iterate on this large model, hoping to add more content and make it more interesting.

Thank you for your support!

------------------------------------------------------------------------------------------------------

推荐参数

Recommended Parameters:

Sampler: DPM++ 2M Karras alt Karras or DPM++ SDE Karras

Steps: 20~40

Hires upscaler:4x-UltraSharp

Hires upscale: 2

Hires steps: 15

Denoising strength: 0.2~0.5

CFG scale: 6-8

clip skip 2

Hello friend👋
I created this mix because I wanted to make a style something between anime and realism and 3d, what happened in the end you can evaluate yourself, there may be mistakes, this is my first experience of uploading models on this site, I hope for your support and also reviews, I will be glad to each of your comments

The model is able to work with both NSFW and censorship. You can get good results on both short and long promts.

Recommend: DPM++2M Karras, Steps: 25-35+
I am using this vae-ft-mse-840000-ema-pruned

For better quality as in the first two examples, you need to use 150 steps and upscale in the image2image tab, 4k-ultra sharp. with these settings you will achieve the best quality, but at 35 steps the quality is also excellent, if you also do an upscale, it will be slightly worse than at 150 steps

If you wish, you can subscribe to my telegram group, where I post my work every day, and you can personally observe how this model changes from day to day
https://t.me/neirosea

Have fun with my mix

Model is a result of mix'n'matching different trained or merged checkpoint with little experimentation with discarding/removing some weights.

Main goal was to create model that primary focus is men and content creation related to homosexuality, male relationships and NSFW gay content.

It's experimentation done just to learn before making a proper rained checkpoint.

用一种特别的方式制作的一个特别画风的模型，底模的名字是Shyaku_JX,是我自己训练的，不公布的模型。这个模型的许多素材是我的朋友Tonade提供的，他的主页是https://civitai.com/user/Tonade/models

A model with a special style made in a special way，The name of the bottom model is Shyaku_JX, which is a model I trained myself and is not published.Many of the materials for this model are provided by my friend Tonade, whose homepage is https://civitai.com/user/Tonade/models

female

This is a Mix Mash of Training and merging. First model... Use your own vae... you should have one? Selections based upon Euler a 40 steps no hirez 512x512 profile.

This should be good for general use.

I have poured my soul into this materpiece. This is a mix of over 50 MODELS. If you appreciate my work, kindly consider supporting me:

• https://www.buymeacoffee.com/LyloGuumy

As a creator, I have always been passionate about creating digital art that blends realism with a touch of anime. That's why I'm excited to introduce you to my latest project: IceRealistic, a unique simulation model that generates stunningly realistic images of ice and snow, with a twist - my model features beautifully rendered anime girls that bring a new level of charm and elegance to the art.

IceRealistic is perfect for anyone looking to create game assets, architectural visualizations, or scientific research projects that require realistic renders. With my model, you can create stunningly detailed images that will capture the imagination of your audience, with the added bonus of a touch of anime charm that will make your work stand out from the crowd.

But to continue creating and improving my model, I need your help. Donating to my ko-fi page will allow me to invest more time and resources into my work, allowing me to create even more stunning and realistic images for you to enjoy. As a token of my appreciation, I will provide exclusive content and perks to all of my donors, including early access to new updates and personalized tutorials.

So if you're a fan of anime and love seeing digital art come to life, please consider donating to my ko-fi page. With your support, I can continue to create amazing art that captures the beauty of the natural world, with a touch of anime magic. Thank you for your support!

FAQ:

- Best Resolutions without HiResFix Requirement

I personally believe my model works best at either 512x832 and 576x768

- Recommended Samplers

DPM++ SDE Karras at 28 steps or DM++ 2M at 50 steps

- Recommended Steps

For SDE: 28 steps with 8 highres

For 2M: 50 steps with 15 highres

- Recommended CFG

Anything between 7 and 10.5

- Restore Faces Possible (Yes/No)

If it works for you sure, I personally haven't used it.

- Recommended VAEs

kl-f8-anime2 if you want vibrant colors, alternatively any VAE is good.

- Recommended HiResFixes Samplers

4x-UltraSharp is the best IMO

- Bonus

Look up Detection Detailer for better faces! My favourite~

This is a merge model of a couple models and loras, I can't remember each one but I believe the base was made of Mistoon.

This was going to be a personal model, but the frens on Baest were interested in it, so everyone gets it.

It works well with the Anything VAE, if your generations happen to be bland or colorless or have purple artifacts.

Versions: LORA SD1.5 | LORA SD 2.1-768 | LYCORIS SD2.1-768

JDZ Beta1 - 22 April 2023
Beta1 is released for testing and evaluation. While convergence was reached, there are 230 styles/concepts contained within. This was trained on my original datasets used to create many CKPT / LORA already released on Civit. The Training Data contains combined instructions on making the most from this new versatile model format.

UPDATE - added new "combined-instructions.zip" now includes Instructions.txt and wildcards.zip. Wildcards are now included for use with Dynamic prompts.

JDZ-CaptionList.txt - ~5000 Example prompts used in captioning the 230 sets
JDZ-StyleTriggers.txt - ~230 "Trigger words" used for each style

JDZenkai-GenPrompts2200.txt - ~4200 "GPT2 Generated Prompts" trained on the Caption List

(Above pictured) Wildcard files now included

• Future releases will be trained to Epoch 40 (~200,000 steps)

• The Beta1 model is currently trained to Epoch 19 (~97000 steps.

• There are many ways to use this model as it contained all my CKPT styles in one model.
  

1. You do not need to use Triggers and can prompt normally.

2. You can prompt using words listed in DJZ-Styletriggers.txt to anchor to a specific style, or use multiple triggers with attention weighting to control the output.

3. You can use one or more prompts as shown in the DJZ-CaptionList.txt.

4. You can use the DJZ-CaptionList.txt and/or JDZenkai-GenPrompts2200.txt as a wildcard with Dynamic Prompts, simply place it into the /wildcards/ folder and rename it to your choosing.
   
   I trained a GPT2 Generator which created many variations on the Example Prompts, an example of these are the JDZenkai-GenPrompts2200.txt

This is an ongoing project so expect a LORA version next. It will take some time to create the LORA version due to a difference in the captioning. This is trained using a Parallel Dreambooth Method which prevents concept/style bleeding, but you can still mix by interpolating trained terms/triggers in the prompt.


TLDR there is a missing comma, can you find it?

2.5D Anime model based on AOM3 and Sakimichan style, realistic version is based on ChilloutMix, good at drawing anime characters that have a puffy look it come out with mature body, good with lora, good with NSFW, vae is baked in.

Positive Prompts:
(masterpiece), (best quality), HDR, intricate detail

For girls:

1girl, solo focus, skindentation, perfect face, puffy face, beautiful face, big eyes, puffy eyes, perfect eyes, eyelashes, __skin_color__, __eye_color, __hair__

Or

1girl, solo focus, skindentation, perfect face, puffy face, beautiful face, big eyes, puffy eyes, perfect eyes, eyelashes, __popular_character__

Use body parts for full body shot example: chubby body, narrow waist, big pelvis, wide hips

For Negative Prompts:
EasyNegative, bad-hands-5, (worst quality, low quality:1.4), low contrast, monochrome, text, watermark, thunder, skin imperfection

Make sure you have downloaded the embeddings (EasyNegative, bad-hands-5).

Use nsfw, naked, nude, nipples in the negative if you want clothes on your girls.

Recommanded settings:

• Steps: 25

• Sampler: DPM++ 2M Karras or UniPC

• CFG scale: 7

• Size: 512x768 or 512x512

• Denoising strength: 0.45

• Clip skip: 2

• Hires upscale: 1.5

• Hires upscaler: 4x-UltraSharp (can be found on huggingface.co)

Future improvement will be made to the model.

Known issues:

• Third-party hands: I recommend cropping the photo using the Resize Width in the upscaler to avoid third-party hands.

• Faraway eyes are not good.

Change log:

• v1: first version.