This is a merge of my 3 favorite models/mixes:
MothMix

ReV Animated

Pastel-Mix [Stylized Anime Model]

This merge started as a curious question with other people at civitai discord, "What will happen if i 50/50 merge Rev Animated with Mothmix?" The results were great. So I tried merging Pastel-mix in it too to give that pastely effect but keeping the properties of previous merge. And this is the result. I hope it didn't caught my bad hands curse.

NOTE: This mix is based on my personal tastes.

Recommended config

Sampler: DPM++ 2M Karras

Sampling Steps: 25-40

Upscaler: 4x-UltraSharp

CFG Scale: 5-10 (what I usually use with any model)

Denoising strength: 0.2-0.7

Clip Skip: 2

Vae: None(if you want washed style), Blessed-fix

Enable Quantization in K samplers

The images are not cherry picked, these are simple hires fix 2x upscaled images that are generated while reviewing few chan LoRA that we people at civitai discord server been working on. Although some seeds were repeated with updated prompts to get details like hands, extra limbs correct.

Credits

Author of Rev Animated: s6yx

Author of Pastel-Mix [Stylized Anime Model]: andite

Author of MothMix: bombasticmori

Do check their content and support them for their amazing work.

Recipe

As a software dev, I don't like to make simple things complex. So here are the exact screenshots of the merge window.

NOTE: ReV Animated was first pruned to fp16 since I like to keep them as small as possible.

Step 1:

Step 2:

Easy as that.

License

This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage. The CreativeML OpenRAIL License specifies:

1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content 
2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)

Please read the full license here

Disclaimer

The use of this learning model is entirely at the discretion of the user, and they have the freedom to choose whether or not to create NSFW content. 
This is important to note that the model itself does not contain any explicit or inappropriate imagery that can be easily accessed with a single click. 
The purpose of sharing this model is not to showcase obscene material in a public forum, but rather to provide a tool for users to utilize as they see fit. 
The decision of whether to engage with SFW or NSFW content lies with the user and their own personal preferences.

DEEP DARK MODEL

Negative prompt: (worst quality, low quality:1.4), (bad_prompt_version2:0.8), EasyNegative, badhandv4, text, name, letters, watermark

Steps: 20,

Sampler: DPM++ 2S a Karras,

CFG scale: 4,

Denoising strength: 0.5,

Hires upscale: 2,

Hires steps: 20,

Hires upscaler: R-ESRGAN 4x+ Anime6B

Clip skip: 2,

VAE: kl-f8-anime2.vae.pt

➡️A model I made using some custom training and merge with other popular models. I wanted a model capable of easily generating good consistant 2.5D style images with simple prompts.

🔞NSFW capable.

👩One disadvantage of this model is that it might be less efficient for some specific characters.

📝The recipe is quite complex so I don't remember all the models used. Sorry.

✔️Feel free to use the VAE you want. I personnaly use sd-vae-ft-mse-original, and sometimes Blessed2 VAE.

🛠️Recommended tools : A1111 WebUI upscaler and After Detailer.

Negative prompt: (worst quality, low quality:1.4), (bad_prompt_version2:0.8), EasyNegative, badhandv4, text, name, letters, watermark

Steps: 20,

Sampler: DPM++ 2S a Karras,

CFG scale: 7,

Denoising strength: 0.5,

Hires upscale: 2,

Hires steps: 20,

Hires upscaler: R-ESRGAN 4x+ Anime6B

Clip skip: 2,

VAE: kl-f8-anime2.vae.pt

This is a supermerge of many LoRA I've trained for the past few months together with 20% of PhotoRealism 2.0 for stability. It doesn't have any triggers as such but responds well to Large Areola, Cellulite, Post-Partum Belly, BBW and other plus-sized body characteristics. Because Photorealism is included as well as the VAE being baked in it should work well in a number of different uses. I'm looking forward to seeing your results.

CBDMix is a mixed model of BRAv5 and chilled remix V2, ranging from 10% to 90%. Since both of them are released under the openrail license family, the same CreativeML Open RAIL-M license will be applied to the model as well.

My personal recommendation is 60% or 70%, but it's up to you :)

Here are the comparisons, just in case.

chubasart

https://huggingface.co/x1101/models/resolve/main/anymix_v2.ckpt

This model is anything5 with mixed other nsfw model, so it can produce nsfw while retaining the clear flat anime-style,

The model works better with loras, now i use it to generate nsfw stuff on my pixiv page. https://pixiv.me/x1101_id to see more example. all example image using this model

This is a global repository of the Dreambooth Checkpoints.

Some people asked me to upload the source Dreambooths from which I make the LyCORIS.

To not spam the site/profile with multiple models I decided to do it the same way I intended to handle the Game characters -> a single page where each model is just a separate version.

Upcoming: Caity Lotz, Alissa White-Gluz

(if there are no more suggestions from the community, I will be picking myself who to upload next :P)

NOTE1: I will be uploading up to one Dreambooth per day (the visibility of other models suffer and I want all to have an equal chance :P, also this gives me more time for LyCORIS)

NOTE2: Feel free to suggest who should be added (you can look at my existing LORA/LyCORIS library to see which Dreambooth I have; but of course a completely new person can be suggested as well - however that would be best to do via my ko-fi page unless you can wait very very long :P)

NOTE3: This is for those who prefer to get the maximum quality! If you are fine with the quality of the LyCORIS, you may want to skip it (however, I imagine I might upload some models here before I convert them into LyCORIS for time reasons).

However, If you prefer LyCORIS - you need not download anything from here because I eventually will upload all those models in the LyCORIS version :)

model is made based on photos from niji and model of niji3Dstyle and some other models.
The Cute version will have a more petite and lovely style than the Milf version

V2.3.5 Continue to improve limbs and body, eyes are less error but less bright, improved material more like Niji but better light than Niji. I still try to keep what worked well in the previous version but add some things that the old version did not have, also improve things in the old version that is still not very good.

If you like my content, support me at:
https://www.subscribestar.com/duchaiten

Model: DucHaitenNiji-cute

DPM++ SDE Karras is cooler, but DPM++ 2M Karras makes the face looks cuter

CFG 7.5-10

steps 20

N prompt:

(worst quality:1.5), (low quality:1.5), (normal quality:1.5), lowres, bad anatomy, bad hands, vaginas in breasts, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyeblows, (cropped), oversaturated, extra limb, missing limbs, deformed hands, long neck, long body, imperfect, (bad hands), signature, watermark, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, painting by bad-artist

*Or use this prompt, but it will give unstable results, sometimes very good, sometimes bad

illustration, 3d, 2d, painting, cartoons, sketch, (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, bad hands, vaginas in breasts, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyeblows, (cropped), oversaturated, extra limb, missing limbs, deformed hands, long neck, long body, imperfect, (bad hands), signature, watermark, username, artist name, conjoined fingers, deformed fingers, ugly eyes, imperfect eyes, skewed eyes, unnatural face, unnatural body, error, bad image, bad photo

• 擅长卡通场景，说是MIX模型，其实加入了多个已发表或私藏的训练LORA。

• 适合简单的关键词，如：masterpiece,best quality,mini landscape

• 由于主要用来出场景，所以负面关键词不用加

• 关键词推荐：mini landscape，indoor, toon，pixar， scene，building，vehicle，

-- -- -- -- --

• used for cartoon scene,contain some my trained lora

• try simple prompt,like : masterpiece,best quality,mini landscape

• no need neg prompt

• useful prompt ：mini landscape，indoor,toon，pixar， scene，building，vehicle，

Trigger Word: "gdmint luxury modern interior design"

This is the ultimate Luxury Modern Interior Design Checkpoint trained on the style of GDM with 26k steps, 260 images, 3k regularization.

RECOMMENDED SETTINGS SD 1.5:

Steps: 60

CFG: 8.5

Trigger Word: "gdmint luxury modern interior design"

Tip#1: Try to use the above Steps and CFG settings first and tweak from there. I think 40 steps may be the minimum acceptable quality. CFG 7 is minimum and after 10, will be a hit or a miss.

Tip#2: Using prompt "gdmint luxury modern interior design" works best to trigger the look, then add your own prompt to spice up the picture! However, if you want to use the raw mode, use only "gdmint" and it will reference those trained pictures without any addons.

As a favor, please let me know your thoughts, feedback and give a review rating. I would truly appreciate it!

Colorful comic style model, made by megering many loras and models, good at creating cute anime girls. Hope you like it.

credit:

Models: mikapikazo, crsb, wonton-colorbox-enhanced,

Loras: Gashi-Gashi style, akai_sashimi style, Nukunuku Nigirimeshi style, argento style, Squeezer2

IF YOU ENJOY THE MODEL PLEASE RATE/REVIEW IT!!

I found that most models are to opinionated, hence the multi model. It excels at detailed prompts and doesn't try to turn everything into a human (but you still can generate the human's ofcourse).

For reference please check the images posted, for the first few I kept all the settings the same so to get the result all you have to do is copy the pos & neg prompt and you're done. The ones after that i just copied from other posts and posted the ones that came out. Thanks for those prompts.

I seem to run on coffee so if you want to keep me running, here is my coffee link (also taking custom requests in exchange for for a couple of double espressos, slide into dm): https://www.buymeacoffee.com/olliesd

recommended settings are:
DPM++ 2M Karras
CLIP skip 2
for detailed prompts use high CFG.
Hires fix makes everything better!
no vea required

FYI if you're getting image collections instead of 1 images try tuning down the weights.

When prompting for people it'll tend to look like chilloutmix with some softer lighting. Although the same thing can easily be achieved by adding some shadow & skin related tags to your prompt.

This model is also surprisingly good at inpainting!

GLHF & IF YOU GENERATE ANY AWESOME IMAGES, SHARE THEM PLEASE :)

概要/About

NAIリークよりシャープで若干鮮やかかつWDのようににじまないマージVAEです。

This is a merged VAE that is sharper and slightly more vivid than the NAI leak and does not bleed like kl-f8-anime2.

アニメ調モデル向けに作成しました。リアル調モデルでの使用は想定していません。

Made for anime style models. Not suitable for realistic models.

V2はかなりシャープな絵になる代わりにざらざらした質感になります。気になる方はNansLess1かV1.0をご利用ください。

マージに使ったすべてのVAEが変色かにじみがあるためシャープネスと変色抑制の両立はできませんでした。ご了承ください。

注意/Note

V1はNAIの影響でNansException(黒画像)が発生する場合あります。

1111のコマンドライン引数に--no-half-vae(速度低下を引き起こす)か、--disable-nan-check(黒画像が出力される場合がある)を追加してみてください。

V1 has frequent Nans errors due to NAI.

If you see a NansException error, Try add --no-half-vae(causes slowdown) or --disable-nan-check(may generate black images) to the commandline arguments.

エンコーダーは調整していません。学習には適していません。

Encoder is not adjusted. not suitable for learning.

MeinaMix objective is to be able to do good art with little prompting.

I created a discord server where you can post images that you generated, discuss prompt and/or ask for help. https://discord.gg/meinaverse

You may also try my model using Sinkin.ai: https://sinkin.ai/m/vln8Nwr

If you like one of my models and want to support their updates:
I have a ko-fi and Patreon page where you can support me or buy me a coffee <3 , it will be very much appreciated: https://ko-fi.com/meina and https://www.patreon.com/MeinaMix

MeinaMix and the other of Meinas will ALWAYS be FREE.

The images in the samples aren't cherry picked, if i had spent 2h to cherry pick each of the images it would hide how the model really performs and i don't think thats fair.

Recommendations of use:
--------------------------------------------------------------------------------
Enable Quantization in K samplers.

Hires.fix is needed for prompts where the character is far away in order to make decent images, it drastically improve the quality of face and eyes!
---------------------------------------------
Recommended parameters:
Sampler: Euler a: 40 to 60 steps.
Sampler: DPM++ SDE Karras: 30 to 60 steps.
Sampler: DPM++ 2M Karras: 20 to 60 steps.
CFG Scale: 4 to 11.
Resolutions: 512x768, 512x1024 for Portrait!
Resolutions: 768x512, 1024x512, 1536x512 for Landscape!
Hires.fix: R-ESRGAN 4x+Anime6b, with 15 steps at 0.4 denoising.
Clip Skip: 2.
Negatives: ' (worst quality, low quality:1.4), (zombie, sketch, interlocked fingers, comic), '
--------------------------------------------------------------------------------

In the merged models list: MeinaMix V1~6, MeinaPastel V3~5, MeinaHentai V2~3, Night Sky YOZORA Style Model, PastelMix and Facebomb, i do not have the exact recipe because i did multiple mixings using block weighted merges with multiple settings and kept the better version of each merge.

☕https://www.buymeacoffee.com/Sniteur - If you like my work ❤

V2 - Training... 5% - This version is preferred for portraits only.

• Improved visual rendering on higher resolutions.

• Improved light effects and lighting.

• Improved light-in-dark play.

• Removed repetitions and improved model size.

V1.1 - Training... 0.7%

• More photorealism & and more eye and light details.

V1 - Training... 0.5%

• ... You have to start somewhere.

Recommended configuration:

• Clip skip : 2

• Sampling steps : 20 ~ 50

• Sampling method : DPM++ 2M Karras / DPM++ SDE Karras / DDIM / Euler / Euler A

• CFG Scale : 7 ~ 12

Very Excited to Release 3.0! Fine tuned it a bit more. Robust and versatile model that still has a variety of capabilities creating characters. I enhanced portraits, landscapes and realism.

Follow on twitter, requests please DM at https://twitter.com/fitCorderAI

Characters: Tsunade, Orihime, Beerus, Fasha, Enhanced Game girls

Body Type: Double Thick Thighs, Perky Breasts, Round Breasts, Max Cleavage, Amazon Goddess, Big ol butts

Hair: The Karen, Cybergoth

Misc: Girls on hoods, Nerdy, Futurescape, Undersea, 2D Pixel art

Fashion: Crop Tops, bodycon minidress, Scrunchbutt leggings, Cybergoth, Nerdy, bodycon skirts, bodycon shorts, Tsunade Kimono, Flared Pants, Slingshot Bikini, Saiyan Armor, Moxxi, Swim cover, Titan Leggings, Bicycle outfit

Here is my base model. FitCorderMix. My realistic mix is fCBlendMix. My anime mix is fCAnimeMix.

This is a merge of my previous model Disney Pixar Cartoon Type A with my own lora trained on midjourney images. Compared to previous version, the males and old people look more like actual pixar character and females looks less anime.

I Use hires. fix, upscale latent, denoising 0.5, hires steps 20, upscale by 2 . Clip skip 2 .

I use night sky yozora VAE in the examples https://civitai.com/api/download/models/14459?type=VAE

if you like the models, please consider supporting me on, i will continue to upload more cool stuffs in the future
patreon.com/PromptSharingSamaritan
https://ko-fi.com/promptsharingsamaritan

IF YOU'RE NEW HERE

U need to use a proper VAE or the color become pale, and it will look way better if you use hi-res fix

copy the yozora VAE to stable-diffusion-webui-master\models\VAE and then go to automatic1111 webui , click on settings tab, the go to stable diffusion tab to the left, then go to SD VAE to select which VAE u want and then click apply settings.

****** WARN: This is not for normal users ******

This is the chillyMixBaseV0 used by chillySauceMix.

If you are not interested in checkpoint merge or development, you don't have to download it.

• This checkpoint alone can produce too many overfitting results. such as

  • resulting images are too much exaggerated

  • torn or mixed broken

  • these images might be harmful to children or pregnant women.

Merge Recipe

please see chillySauceMix(by @mixboy) description

Why use it?

If you're interested in developing checkpoint files, you might be wondering how a lot of checkpoint files were created, or what the original source of the merged checkpoint files was.

However, many checkpoints are being merged and distributed, either not properly identifying what the original source is, or ignoring the licenses of the original source. This can cause potential licensing problems, and there are checkpoints that risk being retired due to licensing issues in the near future.

To resolve these issues, it is very important to use freely available checkpoints with a free/open license.

Interestingly, a bad checkpoint file + a bad checkpoint file + a bad checkpoint file and so on can get good results. so you can merge this bad checkpoint as one of your basis models.

License

Like as ChillyMix, its license has no commercial restriction. it is open and free under CreativeML-Open Rail-M license:

This model permits users to:

✔Use the model without crediting the creator

✔Sell images they generate

✔Run on services that generate images for money

✔Share merges using this model

✔Sell this model or merges using this model

✔Have different permissions when sharing merges

Donation

Feel free to donate to me if you want to support my work! at ko-fi https://ko-fi.com/mixboy

Valkyrie_Mix is meant to be a beautiful and colorful/bright anime and evnironmental mix. This is meant for SFW but NSFW is possible!

Lets just say I needed a bit more awesomeness in my life, well here we are, something to satisfy my need for greatness, any maybe yours too!

-

This is a custom creation! (with the help of others!)

-

Secret VAE!?! : O

-

It is recomended to use (512x768), (Euler A), and a (CFG Scale of 7)

(768x768) images are still good and possible :

• Example Generation

Positive prompt: (extremely_detailed), The Final Stand

Negative prompt: (worst quality, low quality:1.4), easynegative, bad anatomy, bad hands, cropped, missing fingers, missing toes, deformed, disfigured, distorted face, mutation, ugly, poorly drawn hands, out of focus, monochrome, symbol, text, logo, door frame, window frame, mirror frame, elf

Steps: 30, Sampler: Euler a, CFG scale: 7, Seed: -1, Size: 768x768,

-

Thank you the creators and mods of Yodayo, AI generation website, as without them i would not be here today making good images and checkpoint merges.

-

Special thanks to you, for downloading! : D

This is DynaVision, a new merge based off a private model mix I've been using for the past few months. The output is kind of like stylized rendered anime-ish... I guess? I don't know how to classify it, I just know I really like it, and everybody I've let use it really likes it too, and it's unique enough and easy enough to use that I figured I'd share it with the community. It's got noise offset mixed in, so it should be able to create nice dark scenes easily, and really bright ones too. Also, I've found that it's very responsive to Danbooru + clip skip 2. It has the absolutely incredible Detail Tweaker LORA baked in, but it still responds well to adding EVEN MOAR DETAILS with the LORA I've found so go nuts! 🥜🥜

For the samples I'm using my own embeddings, you can find them over on my Civit page, along with a few new ones that I'll be uploading soon to my toolbox. I hope you like this model, I built it with the intention of producing images that are less photorealistic and more artistic stylized output (seriously, it's in-progress name was "ArtsyFartsy").

No face inpainting, no ControlNet, no input masks, all of the samples you see are raw, the only thing I use is High Res Fix for most (though not all). I get best results using DPM++ SDE Karras - Check the gen info on all of my sample images, it's all there. Enjoy! 💖

Model merge using OS-Honey and Gold (v1) and backcrossing it with Honeydew.

https://civitai.com/models/74688/os-honey-and-gold
https://huggingface.co/BakingBeans/Honeydew/resolve/main/Models/Honeydew.safetensors

This merge is made to give a more realistic style to Honey and Gold, while it looses a bit of its magic feel, it gains more control and keeps the beautiful landscapes,

this is a bit heavier of a model (~4gb) but it is much more versatile and requires less/no negative prompting compared to the H&G v1 and v2 models.

this model has more of a "realistic" anime style,
if you feel you are getting grayish/gloomy images, that can maybe be fixed using a vae, or using (blurry:1.5-2.0) in the negative prompt(idk why this works). this happens sometimes because honeydew does not have a VAE Baked into it.

This is a free/open available merged checkpoint model.

All models have merged with the SuperMerger extension

• chillyMixBaseV0 described in the chillySauceMix by @mixboy

• anything v3.2 at Hugging Face

so this model includes well-known legacy checkpoints as follows:

• pyro's_pov models - (pyro_pov_a + pyro_pov_b)

• F222 (Fp32) see also F222 at https://rentry.co/sdmodels

• r34_e4 (Fp16) - from https://rentry.co/sdmodels

• and anything v3.2 at Hugging Face

Merge Recipe

by SuperMerger extension following two-step merge can be done in one step:

• (chillySauceBaseV0 x 0.8 +anything-v3-2-Fp16Fixed x 0.2) x 0.4 + neverendingDreamNED_noVae-Fix x 0.6

the ChillySauceBaseV0 model alone is not good enough to use alone. It produces overfitted results sometimes. adding anything-v3 reduces this issue significantly and add NED(NeverEndingDream - NO VAE version) again to make it more robust and good enough for normal use.

• overall anything-v3-2 contribution = 0.2 x 0.4 = 0.08 (anime style)

• overall r34 contribution = 0.25 x 0.4 x 0.8 = 0.08 (anime style)

What's the difference

and what are the advantages of this unlike anything other models?

• first, its recipe is open, anyone who wants to change the results a bit can change their recipe based on it.

• more realistic body/skin

• less realistic face (2.5D style): r32_e4 + anything-v3.2 contribute to it.

• less Asian style - you can use the PureErosFace(v1) embedding for Asian women.

What's the difference between chillySauceMix

• sweetChillyMix includes the Anything-v3.2 :)

• famous well-known legacy checkpoints are included.

• basically, chillySauceMix is for checkpoint developer/merger but SweetChillyMix is for normal usage.

Recommends

• default Clip skip is 2, ETA noise seed delta: 31337

• chilloutmix based LoRA works fine.

Useful AUTOMATIC1111's webui extensions

• SuperMerger extension

• Lora Block Weight extension

• webui model toolkit to examine Loras and models.

• civitai extension to add several useful features for civitai.

• ADetailer or DDetailer to enhance face details.

License

Like as ChillyMix, its license has no commercial restriction. it is open and free under CreativeML-Open Rail-M license:

This model permits users to:

✔Use the model without crediting the creator

✔Sell images they generate

✔Run on services that generate images for money

✔Share merges using this model

✔Sell this model or merges using this model

✔Have different permissions when sharing merges

Donation

Feel free to donate to me if you want to support my work! at ko-fi https://ko-fi.com/mixboy

Replicates the realistic art style of Hungrypaws gay anthro male art.

My Sensei's Model

He fused Legacy (My Model) & Darksushi 2.5_v1, with a bit of RuviMix in the mix.

Descendant to Something_Amazing

I will be working with this model more.

The progress I've been making this is pretty nice, if there any suggestion guys of what other model's should I merge it with?

My Very Own Model

Just a small fusion of Somethingv2 and RuviMixV1, still a working progress but I will someday make it work

This model is the result of a merging of Counterfeit and Hassaku. The model has a bright and beautiful anime style with high detail.

Using model:

Use Counterfeit 2.5 VAE : https://huggingface.co/gsdf/Counterfeit-V2.5/resolve/main/Counterfeit-V2.5.vae.pt

Textual Inversion: EasyNegative, bad-hads-5, Deep Negative, bad-artist-anime.

For high detalization use Latent upscale with 0.6 denoithing.