A mix of models I like, I combined this to try and get a better look for my Loras.
I general gives very good images in my opinion.
RevAnimated
Kotosmix
DarelitesFantasyMix
I recommend using DPM++ 2M Karras as the sampler
If you want realism simply add (Realistic:1.5)
Have fun!
If there's anything you need, please don't hesitate to let me know.
This model was initially intended to produce cute anime characters, but it has also been discovered that it has the capability to produce Softcore NSFW content.
Version 2 can produce good results with fewer steps, but the number of steps can affect the final image composition.
here is a link to access the model on Hugging Face: https://huggingface.co/Kybalico/CalicoMix.
I merged models and created example images using my poor laptop with a GeForce GTX 1650 (4GB memory).
If you like this model, you can support me at https://www.buymeacoffee.com/kybalico.
Negatives: EasyNegative, (verybadimagenegative_v1.3), (missing fingers, missing hands, missing legs:1.4) (extra fingers, extra hands, extra legs:1.4), (mutate fingers, mutated hands, mutated legs:1.4), (blurry, depth of field:2.0), (poorly drawn hands, poorly drawn face:1.4), signature, watermark
Sampler: DPM++ SDE Karras
Sampling steps: 20+ (recommended 30-40+)
Resolution:
image compose change by resolution
384x512 (available, but may result in less detail)
512x768 (recommended)
768x1152 or above (for more details)
For landscape images, reverse the width and height values accordingly.
CFG Scale: 7
Clip skip: 1-2
Hires.fix: R-ESRGAN 4x+Anime6b
Denoising strength: 0.33
Version 2:
improved background and enhanced details, modified style
Version 2.5:
fine tuning
"Clam" is named from using modes' name.
It has better performance in txt2img when using positive prompts to describe what you want without any quality tags, and you can use EasyNegative as negative prompt.
Ver.1.1 has more details (like background,body...).
Ver.1.2 has more anime-styled hands (but it may generate strange figures).
hf:https://huggingface.co/Johnhex/Clam
Here is the parameter recommended to use in stable diffusion.
VAE: animevae.pt [f458b5c6]
Sampler: DPM++ 2M Karras or DPM++ 2S a Karras
Step: 15~30
Hires Upscale: 2,Latent Denoising:0.5~0.8
Hires Upscale: 2,R-ESRGAN 4x+ Anime6B Denoising:0.2~0.6
CFG Scale: 7~15
Sorry that I forgot to record using models' weight.
Color Box Model [a80fee84]
Ligne Claire Anime Diffusion V1 [ee5b91da]
Anything-V3.0-pruned [2700c435]
tmndMix-tmndMixPlus [65b8d0b2]
momoko-e [a2a802b2]
v1.1 uses v1.0 and models below:
bluemoonmix_v1.safetensors [b684ed630c]
cuteyukimix_v20.safetensors [7f15df40ed]
lineAndLight_v10.ckpt [f627448629]
v1.2 uses v1.1 and models below:
cuteyukimix_v20.safetensors [7f15df40ed]
dualPersonality_dualLinnetree.safetensors [3e1f22b617]
A stylized version of RuviMix!
masterpiece, best quality, 1girl, mcnsfw, (princess peach:1.2), mini crown, long dress, blonde hair, pink dress, arms behind back, smile, wide-eyed, open mouth, :d, happy, outdoors, castle, balcony, scenery, forest, mountainous horizon, sky, looking at viewer, cowboy shot <lora:mcnsfw_3597:1>
Negative prompt: blurry, (low quality, worst quality:1.3), (loli, petite, aged down, child:1.2), watermark, artist name, censored
Size: 512x768, Seed: 3757621237, Steps: 20, Sampler: Euler a, CFG scale: 7, Clip skip: 2, Model hash: f303d10812
This model merges with my own 3 models called R-base (Realistic), F-base (Fantasy), and V-base (Vibrant)
*** This Model needs VAE !!! ***
The purpose of this model is to generate epic fantasy images that are not only realistic but Animation also.
These models require a VAE. You can try mse-840000-ema or anything. I have a photo of the comparison in the Comparison section (below)
If you got bugs, adding '--no-half-vae' to the argument will solve it.
You use it with any steps, prompt, sampler, ... (anything you want)
Prompt: If you don't know what to prompt, Fantasy Prompts by TxcTrtl is a great choice.
Steps: I recommend 30-45 (You can use higher, but many times do not have a big difference)
Sampling Methods: You can see the comparison in the Comparison section (below).
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)
v1: Pilot version. Aim to be realistic (Cinematic style, Realistic photo, less fantasy)
v2: All tunned. It's stable since version 2.1
[Next update] will be a major change, adding C-Base (Composition) to merge within.
# VAE
# Versions
# Steps
# Sampling Methods (with steps 45)
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)
Pencil drawing of girls wearing glasses. I let the AI learn my drawings.
Initially built to create lifelike portraits of characters from the novel I'm working on, whose title is Endless, thus the name of the model.
This is a rather complex merge, and sadly I did not took notes.
Quick discussion about prompting
It reacts pretty well to (worst quality, low quality, normal quality:1.7) in the negative prompt and this seems to be responsible for the sharp natural texture of the skin, as well as some of the lighting/shadows.
easynegative and NG_DeepNegative_V1_75T mainly have a stylistic effect.
Repeating some keywords in the prompts is sometimes a good idea. For example : realistic, best quality, photo-realistic, (best quality:1.3), realistic, photo-realistic can give better results than simply (best quality:1.3), realistic, photo-realistic.
NSFW content is possible, nsfw, uncensored in the negative prompt can help get rid of them.
I will be working with this model more
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: Natalie Dormer, Billie Eilish, Eva Green, Olivia Wilde
(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 :)
sks woman
05-14 : v4.0
This model has tried to change so that both Westerners and Asians can express it well.
In addition, we tried to make dark expression more active.
As a result, the overall tone of the person comes out dark, but it looks better on a dark background.
* The example photos I upload are just examples, and SD environments are all different, so even if you use the same prompt, you may get different results.
Please see the image as an example only. *
Make good use of light-related prompts.
05-01 : v3.0
In version v2.3, the anime tendency was slightly visible, so I tried to make it more realistic again.
04-24 : v2.3
This version has changed mixing ratios compared to the previous v2.2 and includes some experimental attempts. As always, extensive testing has not been conducted.
I uploaded the pruned version separately, but it is currently downloading only the full version maybe there is a problem with Civitai. I will solve the problem and upload the pruned version within the 25th
Upload pruned model finished
04-10 : v2.2
I attempted to improve the skin tones and lighting and shadows, which were previously awkward. I have noticed a difference in the image, but I have not had the opportunity to thoroughly test the changes due to my limited skills.
VAE : vae-ft-mse-840000-ema-pruned
Steps: 50, Sampler: DPM++ SDE Karras, CFG scale: 20, Size: 512x768, Model hash: a8907e32c0, Denoising strength: 0.35, Clip skip: 2, Hires upscale: 2, Hires steps: 20, Hires upscaler: R-ESRGAN 4x+, Dynamic thresholding enabled: True, Mimic scale: 10, Threshold percentile: 99.95, Mimic mode: Cosine Up, Mimic scale minimum: 0, CFG mode: Half Cosine Down, CFG scale minimum: 15, Eta: 0.2, DDetailer prompt: "", DDetailer neg prompt: "", DDetailer model a: "bbox\mmdet_anime-face_yolov3.pth [51e1af4a]", DDetailer conf a: 30, DDetailer dilation a: 4, DDetailer offset x a: 0, DDetailer offset y a: 0, DDetailer preprocess b: False, DDetailer bitwise: None, DDetailer model b: "None", DDetailer conf b: 30, DDetailer dilation b: 4, DDetailer offset x b: 0, DDetailer offset y b: 0, DDetailer mask blur: 4, DDetailer denoising: 0.35, DDetailer inpaint full: True, DDetailer inpaint padding: 32, DDetailer cfg: 11
04-01 : v2.0
As always, I didn't conduct many tests. However, I did notice a slight improvement in prompt recognition.
I'm not entirely certain if it's an actual upgrade from the v1 version or not.
Steps: 40
Sampler: DPM++ SDE Karras
CFG scale: 20
HIRES. FIX
R-ESRGAN 4X+
STEP : 20
DENOISING : 0.35
UPSCALE : 2
Dynamic Thresholding
MIMIC CFG SCALE : 10
PERCENTILE : 99.95 OR 99.9
MIMIC SCHEDULER : COSINE UP
MINIMUM VALUE OF MIMIC : 0
CFG SCHEDULER : HALF COSINE UP
MINIMUM VALUE OF CFG : 0
Ddetailer
bbox\mmdet_anime-face_yolov3.pth
Denoising strength(Inpaint) : 0.35
Mimic CFG Scale : 15 (custom ddetailer setting)
----------------------------------------------------------------------------------
03-17 : v1.0
I have updated the model with the extension "safetensors" upon the request of someone else, while keeping it the same as the existing model.
----------------------------------------------------------------------------------
This is a realistic model that I personally merged.
There is limited information due to a small number of tests conducted.
This is an example of a prompt used
ultra realistic details, sharp focus, detailed skin, (21 years old pretty korean girl:1.2), (kpop idol:1.2), (korean beauty:0.8), (korean mixed),
(large breasts),
Negative prompt: Drawings, abstract art, cartoons, surrealist painting, conceptual drawing, graphics, (low resolution:1.4), (blurry:1.3), (strabismus:1.1),
(worst quality:1.3), (low quality:1.3), (thick thighs:1.2), collage, (makeup:0.8), nsfw, bad proportions, earings, floral print, loli, big eyes, (watermark:1.2), letter, (abs:1.2),
Steps: 40, Sampler: DPM++ SDE Karras, CFG scale: 7.5, Seed: 3203463706, Size: 768x512, Model hash: 1d8d6e1204, Denoising strength: 0.35, Clip skip: 2, Eta: 0.2
HIRES. FIX
R-ESRGAN 4X+
STEP : 20
DENOISING : 0.35
UPSCALE : 2
Dynamic Thresholding
CFG : 20
MIMIC CFG SCALE : 5
PERCENTILE : 99.95 OR 99.9
MIMIC SCHEDULER : COSINE UP
MINIMUM VALUE OF MIMIC : 0
CFG SCHEDULER : HALF COSINE DOWN
MINIMUM VALUE OF CFG : 10
I made a neat 2D model with animation style. As a model suitable for detailed expression, it can do good art design.
The haku3490 style model was trained on 529 images and has a strong NSFW style. Its version 2.0 reached the desired 70% accuracy level.
I am merging models that I use a lot.
recipi
35% DivineEleganceMix https://civitai.com/models/6174/divineelegancemix
35% Alstroemeria Mix https://civitai.com/models/38606/alstroemeria-mix?modelVersionId=44530
30% MeinaMix https://civitai.com/models/7240/meinamix
I hate Widow player. plz don't 1tap
I don't know why, but this model becomes clearer or simpler depending on the prompt.
Model mix aims to create the most realistic and natural images possible. It's currently in the testing process, so please comment.
Please support me by becoming a patron:
Available on the following websites with GPU acceleration:
Guide:
For the settings or parameters, I recommend using these settings.
Sampler: DPM++ SDE Karras or Ruler a
Steps: 30-50
CFG Scale: 7.5
How to use:
Structure: render for a +<subject>+<details>+<lights>+<color>+<resolution>+<option>
For example:
render for a girl, beautiful face, autumn lights,pastel colors, high quality, trending on ArtStation, trending on CGSociety,(extremely detailed CG unity 8k wallpaper)
Negative Prompt:
illustration, painting, cartoons, sketch, (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, bad hands, ((monochrome)), ((grayscale)), collapsed eyeshadow, multiple eyeblows, vaginas in breasts, (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
Can you customize it all to your liking and show me it? Thank you!!!
LORA is not added yet
HuggingFace:
I started merging vae because I wanted a color that was somewhere between counterfeit and barrymix, and eventually found what I was looking for!
Add --no-half-vae if it fails to work
+In response to feedback I've created a comparison table with other VAEs! If you'd like a slightly larger table, I'll make it a bit larger! (i did not use Dynamic Thresholding to compare all results.) These results are from using the unweighted imputation verybadimagenegative_v1.3. If a different embedding is used, you may see different results
visit me on Twitter -> https://twitter.com/nzxt_cam
[[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
The description has been updated to v3 and v4.
To explain cutemix, this model is specialized for cuteness, especially for portrait bases. For those who value backgrounds and glamor, please use a different model, or merge models with good backgrounds separately!
Unlike other models, cutemix does not use easynegativ, bad prompt, badhandv4, and verybadimagenegative_v1.3 to get the colors you see here.(civitai)
For sampling methods, I recommend unipc, dpm++ 2m karras, dpm++ sde karras, unipc is the best among them.
You can use any upscaler you like, I used boymebob-redux.
(I'll only post FP16 because it's over 10 gigs from v3)
Using my VAE will make it feel even cuter -> https://civitai.com/models/65728/cute-vae
visit me on Twitter -> https://twitter.com/nzxt_cam
Duchaiten-lofi are two models with almost the same style, the A version is less detailed and more anime-oriented, the R version is more complex and a bit realism. Both are not Japanese anime style but will be a bit western.
Please support me by becoming a patron:
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
mbw方式融合,素材模型...忘了QAQ 欢迎大家使用
Welcome to the exciting world of LewdPunk Anima - an innovative stable diffusion model that brings you the best of nude, skimpy clothing styles and futuristic sceneries in realistic 3D or photography style with different lighting modes. Immerse yourself in an exquisite world of beautiful, captivating images and explore the endless possibilities of creativity with our special blend of models. Take a look at our sample images and see for yourself the amazing results we can offer. Try Lewdpunk Anima today and unlock the potential of your imagination!
Visual clarity above all else
This checkpoint is a mix between my own and other checkpoints. It does everything, but excels at close to medium portraits, photorealistic or hyperrealistic. It can even maintain sharp eyes at lower resolutions of 512 x 768, for example. Upscale 2x with hires. fix, and see the eye detail further increase.
If you enjoy the models I create, please consider buying me a coffee!
YabaL_Mix V3
Recommended settings (추천 셋팅)
DPM++ 2M Karras
Sampling steps : 30
Upscaler : R-SERGAN 4x+ Anime6B
Hires steps : 15 to 20
Denoise : 0.4 to 0.6
CFG : 06 to 15
Realistic: Possible. (가능)
Recommended value: 1.0 to 2.0 (권장 값 : 1.0 to 2.0)
NSFW: Possible.
Recomended VAE : vae-ft-mse-840000-ema-pruned
Lora : Possible.
I added fp16 version.
The sample image was created using DDtailer. In the default settings, the value of Mask Blur is set to 10.
Sampling Method
DPM++ SDE Karras
Euler a
DPM++ 2S a
DPM2 a Karras
Sampling Steps
40 (20 ≈ 60)
Restore Faces
Off
Hires Fix
On
Upscaler: Latent (bicubic antialised)
Hires Steps: 20 (20 ≈ 40)
Denoising Strenght: 0.5 (0.4 ≈ 0.7)
Upscale by: 2
CFG Scale
7 (6 ≈ 9)
Clip Skip
1 (1 ≈ 2)
VAE
None (Baked)
You can find more images using this model at: https://pixai.art/
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.
girl sexy hijab
hijab 
Please select the version you like, not the latest.
My miscellaneous models.
If you would like to see the recipe or more detail, feel free to ask :>
recipe missing:
(03191422,03192033,03191454)
I was looking for a model with as realistic skin and lighting as possible, but couldn't find one, so I made my own.
I think I created a model that can express Asian women most beautifully by mixing and testing realistic models countless times.
I shared the model with some of my friends and they all liked it so much that I decided to share it.
We will continue to try and test new models in the future.
I hope many people use it.
(nsfw is included)
https://www.youtube.com/@noble_ai
피부와 조명이 최대한 리얼한 모델을 찾고 있다가 찾을 수 없어서 직접 만들었습니다.
리얼한 모델을 수 없이 믹스하고 테스트하면서 가장 아름답게 아시아 여자를 표현할 수 있는 모델을 만들어낸 것 같습니다.
몇몇 친구들한테 모델을 공유하였는데 다들 너무 좋아해줘서 공유하기를 결정하였습니다.
이후에도 계속 새로운 모델을 시도하고 테스트할 예정입니다.
많은 사람들이 사용해줬으면 좋겠어요.
(nsfw 포함되어 있습니다)
https://www.youtube.com/@noble_ai
肌と照明ができるだけリアルなモデルを探しているが、見つからなかったので自分で作りました。
リアルなモデルを無数にミックスしてテストしながら、最も美しくアジアの女性を表現できるモデルを作り出したようです。
私はいくつかの友人とモデルを共有しましたが、みんなが好きで共有することにしました。
その後も引き続き新しいモデルを試してテストする予定です。
多くの人が使ってほしいです。
（nsfw含まれています）
https://www.youtube.com/@noble_ai
我一直在寻找一个皮肤和灯光尽可能逼真的模型，但找不到，所以我自己制作了一个。
我认为我通过无数次混合和测试现实模型，创建了一个可以最美丽地表达亚洲女性的模型。
我和我的一些朋友分享了这个模型，他们都非常喜欢它，所以我决定分享它。
我们将在未来继续尝试和测试新模型。
我希望有很多人使用它。
（包括 nsfw）
Sharing this model with the community. As it closely resembles Lexica Art V2 or Lexica Aperture v2, this product is an excellent choice.
It's absolutely amazing how we're able to create incredibly realistic faces with only a few simple steps.
If you want to support my work, you can buy me a coffee or tea: https://www.buymeacoffee.com/gswan
Enhance your writing prompt by adding the estilovintedois prefix, as it guarantees a consistent writing style.
"While you're welcome to use this model, please keep in mind that any potential damages or mishaps resulting from its use are not our responsibility..."
"This model was trained on a large amount of high quality images with simple prompts to generate beautiful images without a lot of prompt engineering."
"It should also be very dreamboothable, being able to generate high fidelity faces with a little amount of steps."
This model is based on vintedois_diffusion. Credit to the original developers.
Vintedois (22h) Diffusion model trained by Predogl and piEsposito with open weights, configs and prompts (as it should be).
Example prompts
Prompt: photo of an old man in a jungle, looking at the camera
CFG Scale: 7.5
Scheduler: diffusers.EulerAncestralDiscreteScheduler
Steps: 30
Seed: 44
Prompt: kneeling cat knight, portrait, finely detailed armor, intricate design, silver, silk, cinematic lighting, 4k
CFG Scale: 7.5
Scheduler: diffusers.EulerAncestralDiscreteScheduler
Steps: 50
Seed: 44
Prompt: a beautiful girl In front of the cabin, the country, by Artgerm Lau and Krenz Cushart，hyperdetailed, trending on artstation, trending on deviantart
CFG Scale: 7.5
Scheduler: diffusers.EulerAncestralDiscreteScheduler
Steps: 50
Seed: 44
Prompt: destroyed city
CFG Scale: 7.5
Scheduler: diffusers.EulerAncestralDiscreteScheduler
Steps: 50
Seed: 44
Prompt: victorian city landscape
CFG Scale: 7.5
Scheduler: diffusers.EulerAncestralDiscreteScheduler
Steps: 50
Seed: 44
Prompt: prehistoric native living room
CFG Scale: 7.5
Scheduler: diffusers.EulerAncestralDiscreteScheduler
Steps: 50
Seed: 44
estilovintedois
这个模型是我混合了多的模型出来的
底模是葡萄柚
这个比较容易出猫耳（在关键词少的情况下）
可以适应各种风格
并且在nsfw表现良好
但是手部容易出错
Seraph_Mix is meant to be a beautiful and colorful/bright hentai mix.
When I was creating this, I really liked 2 models in specific, but wondered what happened if i mixed them both, and honestly, I was not disappointed. It is really colorful, and amazing and doesn't even require much describing for it to be amazing.
-
Checkpoints involved in Seraph_mixV1 :
Checkpoints involved in Seraph_mixV2 :
Checkpoints involved in Seraph_mixV3 :
Version 4 and 5 are my secret!
baked in :
-
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: 1girl[mature woman], [extremely detailed:1], [medium-small boobs:0.2], [blush, black hair, medium-long hair], bedroom, [sexy black panties, sexy black bra], nice hands, perfect hands], <lora:GoodHands-beta2:1>
Negative prompt: (worst quality, low quality:1.4), EasyNegative, bad anatomy, bad hands, cropped, missing fingers, missing toes, too many toes, too many fingers, missing arms, long neck, Humpbacked, deformed, disfigured, poorly drawn face, distorted face, mutation, mutated, extra limb, ugly, poorly drawn hands, missing limb, floating limbs, disconnected limbs, malformed hands, out of focus, long body, monochrome, symbol, text, logo, door frame, window frame, mirror frame, flat chest,
Steps: 30, Sampler: Euler a, CFG scale: 7, Seed: 2490512948, Size: 768x768,
-
LORA's from "ddp12" are pretty cool.
I also recommend using this LORA : "BetterHands" - By Envy
-
Thank you to the creators of both Yesmix, and AOM3, without them i would not have been able to create such a beautiful checkpoint merge.
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 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.
Just a small fusion of Somethingv2 and RuviMixV1, still a working progress but I will someday make it work
Tennis Player Gamer
ANOTHER MERGE this time trained
50 percent of this models style is due to my 3dmagix lora so enjoy ^^
As always, the example images I've posted have NO postwork done to them beyond the facial fix (in some cases) and upscaling. The point is to show how well it works out of the box - if you take the time and create really good prompts and use good negative prompts (an example is at the end of this post) you can get even better results.
I wanted a model or workflow process that allowed me to quickly and easily churn out art for games and/or photo stories. I didn't want long complicated prompts to create a certain art style, but I also didn't want to be forced to use LoRa's and TI's for every character. Unfortunately, most of the models where I loved the art styles didn't do so well of loading characters (celebrities for mixing, fictional characters, etc.) and the ones that knew tons of famous characters and people tended not take to the art styles I was going for (at least not without complex prompts).
DIGITAL ART
WARLOCKANDBOOBS
CARTOON
MASTERPIECE
<SIZE> BREASTS
<SIZE> ASS
<SIZE> BELLY
HYPERBREASTS
HYPERASS
FLAT CHEST
See "Version Details" to the right for minor revision update notes. This post will be updated with major notes, triggers, and other long-term items.
Rebuilt from the ground up... got rid of some garbage (and pruned it down to half its original size) tweaked some weights that were bothering me, and generally got things cleaned up. I'm not quite happy with the style consistency, yet - but this is a good clean start on things.
NOTE: The trigger words listed are just the ones that were specifically added on top of the SD Base Model - and really only the ones I've found useful to make good tweaks and adjustments. This model is great at knowing art styles, people, and concepts that you might love (and/or hate) from the base model.
I also forgot one ingredient when adding hyperfusion - so it tends to go porny/nude when you start playing with breasts. I won't have time to update/fix until this weekend so... expect to have to do some work to add clothes if you're playing with breast sizes. Sorry about that.
Keywords "Masterpiece" and "add_detail" will enhance (or back down if negative levels are used).
This also responds to the Hyperfusion triggers (see version 3.2 notes) as well as all the custom art styles from Babes. (Plus many many more - but the Babes trained styles are ones likely not to be found in many other models).
A bonus (from Hyperfusion) is that it also lets you accomplish making breasts smaller - something very difficult to do on most models without bringing in a lot of extra help. With this "flat_chest:1.4" (or thereabouts) will do a nice job of shrinking things down in most cases. You can also use all the triggers from hyperfusion to make the breasts, butt, and belly do all sorts of things - and putting in weights there brings varied results too.
NOTE: At low levels, most of those calls don't go crazy by sexing everyone up. The higher your weights, the more likely that is to happen, so you may need to declare SFWs and even call a specific outfit in some cases. (I had to do that in the big bust sample image).
All previous triggers and keywords will still work, too.
It "mostly" defaults to SFW out of the box. i.e. If you just type in a person's name or character description with no clothes - it'll tend to put them in clothes - or at least a swimsuit. That said... with proper prompting, does nudes and porny stuff just fine, too.
I didn't care so much about making sure resulting characters look "exactly" like the person you're calling out so much as wanting to make sure that if I called out that name/person again - they'd be consistent in the way they look from image to image. So, in other words, it's not so much about "accuracy" as it is about "consistency" - and that seems to come through. (This works well for my needs to create stories and whatnot)
Good Negative Prompt for this Model
canvas frame, 3d, ((disfigured)), ((bad art)), ((deformed)),((extra limbs)),((b&w)), blurry, (((duplicate))), ((morbid)), ((mutilated)), [out of frame], extra fingers, mutated hands, ((poorly drawn hands)), ((poorly drawn face)), (((mutation))), (((deformed))), ((ugly)), blurry, ((bad anatomy)), (((bad proportions))), ((extra limbs)), cloned face, (((disfigured))), out of frame, ugly, extra limbs, (bad anatomy), gross proportions, (malformed limbs), ((missing arms)), ((missing legs)), (((extra arms))), (((extra legs))), mutated hands, (fused fingers), (too many fingers), (((long neck))), ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, mutation, mutated, extra limbs, extra legs, extra arms, disfigured, deformed, cross-eye, body out of frame, blurry, bad art, bad anatomy, Scribbles, Low quality, Low rated, Mediocre, 3D rendering, Screenshot, Software, UI, watermark, signature, distorted pupils, distorted eyes, (distorted face), Unnatural anatomy, strange anatomy, things on face, ((tattoo)), ((tattoos)),
Obviously, some of those may want to be removed if you're going for something that has been excluded.
I was happy with my FantasticChix-HR model, so I decided to use it as a base for a more anime-styled version, and these are the results.
Merged models:
FantasticChix-HR
Counterfeit V3.0
Dalcefo Painting V3
Settings are basically the same as FantasticChix-HR, so follow the instructions there for how to use (including the VAE). Since it can sometimes have anatomy problems compared to FantasticChix-HR, I'd recommend also adding a (bad anatomy:1.4) prompt to the negatives.
This model also really likes to generate women, so if you want something else, you'll need to get a bit aggressive with positive and negative prompts. Just using a prompt like "giant mecha" will give you a female wearing mecha armor. Even using just "male" in the positive prompts will sometimes make a more androgynous male.
Only one of the sample images is using a LoRA. The rest are purely the model doing all the work. For some examples of prompts, see the samples of FantasticChix-HR. If you want more colorful results, make sure to include some prompts like "vibrant colors", "colorful", etc.
If you notice anything weird added to the image or any messed up anatomy after upscaling you may need to reduce the denoising strength of the upscaler from 0.5 to 0.4.
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.
Hi,
my another merged model. I am using only neg-anime and plus one two prompts or one more emb for this result. Works well with short prompts too.
Merge
RealisticAI-V6fp16 x -0.4 + pornvision_final x 0.7 + epicrealism_newEra x 0.7
Test merge & mix model
This checkpoint merge from ChikMix_V1 and Orangechillmix_v70 50/50
** recommend to use with a Lora Doll(Korean/Chinese/Japanese/Taiwanese) at low (0.2-0.5) weight and works well with pureeros/ulzzang --- thank you reifonarusief613
Would appreciate images, feedback, requests, reviews, and of course comments in the comments. Please check out my other models if you like this one.
A checkpoint based off NAI trained on 175 Images of various Paizuri.
Later versions will have a larger dataset, the first one was just a test.
paizuri and, paizuri related tags should help, if anyone knows a better model to train of off please let me know. I have not tested using paizuri Loras so don't know if they help or not. can make decent images that aren't paizuri (if you like the way NAI/basic illustrations look)
characters are tagged, and some variants of paizuri, other versions may tag artists and train them as multiple concepts. don't know if Over_clothes_paizuri or Reverse_paizuri were tagged in V1. seems to like to generate simple backgrounds. it is mergeable but don't know how well. a lot of the images are currently flagged sop they won't pop up.
recommended tags: paizuri, cooperative_paizuri
optional tags: pov, from_side, 1girl, 2girls, 1boy, cum, fellatio, oral, unaligned_breasts, symmetrical_docking, asymmetrical_docking, breast_lift, breast_hold, and similar tags
This model uses a new approach to training parallel concepts.
This contains 125 new models + the original 230 contained in the last Zenkai dataset.
(7,500 images trained for ~225,000 steps)
This model will create new images when compared with the Johnson Diffusion Zenkai Model, due to the new approach used.
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.
Update 1.2:
Added fantasticmix_v50 to the merge, now it should be a bit more flexible with fantasy/sci-fi stuff.
Accessories, like necklaces, also are seemingly more consistent.
Updated models in v1.1:
uberRealisticPornMerge_urpmv13
henmixReal_v30
cyberrealistic_v21
revAnimated_v122
deliberate_v2
Focused on chibi art based on artist Derek klaufman.
Making these models can be expensive and very time consuming. If you would consider helping out, it would be much appreciated as any little bit goes a long way
https://www.buymeacoffee.com/Theprotoart
using the regular prompts sometimes results in unexpected outcomes… i’ve tried running it with common prompts, such as manga, comic or vector, anime, etc.
Check out my other models: https://civitai.com/user/mrpbody
evang
Trained on 40 high-quality images from several Kate and Jim McMullen books.
sks style
I used my train result and this LoRA(https://civitai.com/models/13371/fatekaleid-liner-prisma-illya) and OrangeMix, Anything 4.5 etc...
There's no good pixar disney looking cartoon model yet so i decided to make one. This is a merge of Pixar Style Model with my own Loras to create a generic 3d looking western cartoon. In the examples I Use hires. fix, upscale latent, denoising 0.5, hires steps 20, upscale by 2 . Clip skip 2 . I will upload more variants of this in the future
I use night sky yozora VAE in the examples https://civitai.com/api/download/models/14459?type=VAE
if you like the models, you can support me on https://ko-fi.com/promptsharingsamaritan
I don't speak English so I'm translating at DeepL. Let me know if the English is weird.
Multiple SUM merges of the main quality and unique models.
Then I got something that had no personality but was highly applicable to LoRA.
The characteristics of this model are as follows.
Featureless design
Highly adaptable to character LoRA
Highly directional picture with few prompts
Suitable for a wide range of situations
主だった品質の高いモデルや個性的なモデルを複数SUMマージ。
そうしたら個性がないけどLoRAへの適用性が高いものができました。
このモデルの特性は以下です。
個性のない絵柄
キャラLoRAへの適応性が高い
少ないプロンプトで演出性の高い絵が出てくる
幅広いシチュエーションに対応
Models being merged (All of the following models are merged so that they are all in the same proportion)
マージされているモデル (以下のモデルをすべて同じ割合になるようマージしています)
Counterfeit-V3.0 https://civitai.com/models/4468/counterfeit-v30
DucHaiten-StyleLikeMe v1.5 https://civitai.com/models/26069?modelVersionId=47085
Nostalgia-mix https://civitai.com/models/16819/nostalgia-mix
CamelliaMix NSFW v1.1 https://civitai.com/models/44315?modelVersionId=48949
CamelliaMIx 2.5D V2 https://civitai.com/models/44219?modelVersionId=48859
Dark Sushi Mix Brighter Pruned https://civitai.com/models/24779?modelVersionId=33482
Dark Sushi 2.5D v2.0 https://civitai.com/models/48671?modelVersionId=56009
Dorayakimix v3.0 https://civitai.com/models/32355?modelVersionId=57060
CuteYukiMix v3.0 https://civitai.com/models/28169?modelVersionId=68709
GhostMix v1.2 https://civitai.com/models/36520/ghostmix
fantasticmix 2.5D v2.0 https://civitai.com/models/20632/fantasticmix25d
BeenYou A2 https://civitai.com/models/27688?modelVersionId=33228
BeenYou R11 https://civitai.com/models/27688?modelVersionId=64889
MeinaMix V9 https://civitai.com/models/7240/meinamix
TMND-Mix III https://civitai.com/models/27259/tmnd-mix
CarDos Animated v2.5 https://civitai.com/models/22220/cardos-animated
ToonYou beta 1 https://civitai.com/models/30240/toonyou
Sweet-mix v1.7 https://civitai.com/models/18927/sweet-mix
!!VAE is required!!
I used blessed2 for the images but you can use what you want.
https://huggingface.co/NoCrypt/blessed_vae
Forgot password to my old account. Thankfully, I deleted the borked model but I think this one should work now.
Should be able to do very good anime 2.5D illustration style by default. And also accepts artist and style prompts easily. So you can make more 2d or artistic styles.
Merge of some anime and realistic models that I like and have now forgotten which ones made it in.
Check my included images for the diverse prompts. No Loras, textual inversions, or post processing so you can see the model as is. Most are done with hi-res fix 2x, 320x480
If you like it, kindly consider sending a small tip so I can hopefully improve this with a local setup instead of online. 🙏 https://shorturl.at/muNV7
モデルマージはマシンリソースや選定などにコストが掛かっているため、もし気に入って頂けましたらお気持ち程度でかまいませんのでご支援いただけますと幸いです。
Because model merging is costly in terms of machine resources and selection, I would appreciate your support, even if it is only a small contribution, if you like this model.
https://ko-fi.com/aibijo202304
Twitterもフォローしてもらえると嬉しいです。このモデルを使った作品などを投稿しています。
I hope you will also follow me on Twitter. I post works using this model and others.
https://twitter.com/AIbijo202304
ファンタジーの世界もある程度描画できるようにしつつ、どこか日本人らしい雰囲気がある顔立ちの美しい女性が出るように調整しています。
もし使ってみて素敵な作品が出来ましたら共有いただけると幸いです。
I have made adjustments to produce a beautiful woman with a somewhat Japanese look, while also being able to render a fantasy world to some extent.
If you try it and create a wonderful work, we would appreciate it if you would share it with us.
推奨設定
Recommended settings
・Sampling method:DDIM
・Sampling Step:40
・Hires.fix step:40
・denoising strength:0.7
・CFG scale:7
这个模型是私炉的产物，整体风格偏幼化。它可以比较轻松的画出可爱风格的人物，但并非只能画萝莉风格人物，您可以尝试各种有意思的prompt，并且它画风景(背景)效果也很好。
关于v2版本，增强稳定性以及细腻程度，推荐参数如下:Steps: 30, Sampler: DPM++ SDE Karras, CFG scale: 7, Size: 512x768, Denoising strength: 0.57, Clip skip: 2, Hires upscale: 2, Hires steps: 25, Hires upscaler: R-ESRGAN 4x+ Anime6B.
Negative prompt:NG_DeepNegative_V1_75T, EasyNegative, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, (worst quality, low quality:1.4), (bad anatomy), (inaccurate limb:1.2), bad composition, inaccurate eyes, extra digit,fewer digits, (extra arms:1.2), (bad-artist:0.6), bad-image-v2-39000
The reason for this model's name is that it can easily create cute-style images, and its overall color tone is also visually appealing, making it suitable for landscape-style pictures as well. We hope you'll enjoy it!
Regarding the V2 version, to enhance stability and delicacy, the recommended parameters are as follows:
Steps: 30, Sampler: DPM++ SDE Karras, CFG scale: 7, Size: 512x768, Denoising strength: 0.57, Clip skip: 2, Hires upscale: 2, Hires steps: 25, Hires upscaler: R-ESRGAN 4x+ Anime6B.
Negative prompt:NG_DeepNegative_V1_75T, EasyNegative, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, (worst quality, low quality:1.4), (bad anatomy), (inaccurate limb:1.2), bad composition, inaccurate eyes, extra digit,fewer digits, (extra arms:1.2), (bad-artist:0.6), bad-image-v2-39000
v2这个版本改善了远景构图以及远景的梦幻感，里面加入了我经过调试的新的lora-cinematic的参数，以及一些其他的辅助lora，而近景变化不大（手好像没那么稳定了= =）
This version of v2 improves the composition of the vista and the dreaminess of the vista, adding my new debugged Lora-Cinematic parameters, as well as some other auxiliary LoRa, while the close-up scene doesn't change much (the hand seems to be less stable = =)
v3改善了脸部和手部，以及一些肢体结构，还改善了画面精细度。泛用性提高了不少，现在能兼容更多lora了，污染也减少了。
The V3 improves the face and hands, as well as some limb structure, and also improves the fineness of the picture. Versatility has improved a lot, more LoRa is now compatible, and pollution is reduced.
v3.5这个版本改善了偶尔会有的眼睛和脸部的崩坏问题，使得脸部更美型了
This version v3.5 improves the occasional eye and face breakdown, resulting in a more beautiful face
05/15: v1 honestly still looks better than v2 in most cases
25% anything-v5 + 25% HD-22 + 50% Defmix-v2.0 + Baked-In LoRA: 0.25 Detailed Eyes + 0.5 NoiseOffset
Combining the knowledge of Anything & the clarity of Hentai Diffusion & the artstyle of Defmix, while producing beautiful eyes with HDR feels!
50% anything-v3.0 + 50% HD-22 + Baked-In LoRA:0.5 Detailed Eyes
Combining the knowledge of Anything & the artstyle of Hentai Diffusion, while producing beautiful eyes!
Both have the one from HD-22 baked in.
Yes, the name is a wordplay/joke. At least it's not called some randomly-smash-the-keyboard-Mix.
