This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License

the regular creativeml-openrail-m license isn't applicable for this work

Feel free to tag me if you use this model, I'd enjoy seeing what you do with it twitter(@ihtke)

05/14/23 update

see About this version

Queen Marika, 1girl, medium breasts, blonde hair, jewelry, long hair, single braid, solo, yellow eyes, dress, bracelet, black dress, necklace, parted lips, circlet, armlet, pelvic curtain, braid, design by Mark Brooks and Brad Kunkle detailed, gorgeous lighting, wide angle, action dynamic scenery, outdoors

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

well well well....

I'm honorable represent to you my first published creation at the Ai theme, i'll asking you to be kind and do not judge strictly. So first things first lemme tell you a little story just real quick)

I've tamed a beast that made by Kohya(finally that unuser-frendly bunch of scripts have worked just fine.... so far). I spent 4 days to figure it out how works all of it, chose a character, scraped some pics(man... there are not so much of them, especially quality ones), and here i am! After a third run i've got decent(to me) model's result. (yeah this is 4th attempt)

What's this model capable of you might ask?

• Portraits

• Full shots

• and of course, classics cough....

cough...

cough(hysterically).... lewd i've meant, but it highly depends on what model and tools(controlnet etc.) u'll be using and what your power in prompt's knowledge.

About resolution: model's been trained on pics with res. 768x768 so it's mean that not all model will be working properly above that resolution, in my case 768X1024 worked just fine by some tests(yep +1 day wasted on hundreds tests), but then i got encountering anatomical errors and deviations. so go ahead play with it as much as u want🧡

and share with your results cz that will help to improve the current state's model(probably)

Most importantly weights: around 0.9 will be more then enough but you always capable to push through the limits.

For example, for your images, actually just use prompt that shown in the preview or start with these ones: 1girl, jewelry, hair over one eye, dress, bracelet, blonde hair, black dress, long single braid, large breasts, ring, circlet, necklace, armlet, yellow eyes, parted lips, bare legs, single braid, backless dress, hair ornament

P.S. for fuck sake Kohya would u might make that shit much easier to work with, huh? there are fuck 4 days needed to it had worked, fuck...

9th cycle of hell has been passed

The goal of this LoHa (simply called lora below) is to improve the "in container" concept in general

As expected, concept lora is much more difficult to made and is more fragile to model switch

This LoHa is trained on ACertainy with clip skip 1 (but you can use both clip skip 1 and 2; see the the last two example images), and chances are it may not work correctly if you use a model that is far away

In face of this issue, here are some tips for the use of this lora

1. One common strategy is to lower the weight of lora. While this might reduce some weird artifacts, the success rate of getting the concept right also decreases.

2. You can first generate some images on ACertainty with txt2img, and then you use img2img on another model. This is what I did with the refrigerator image. You probably still need to lower lora weight in the second stage.

3. What I do personally is to use some merge that is close to ACertainty. This is done through add difference merge with other model: AC+w(model1-model2)

Do you really need this lora?

In fact, many of these concepts can already be generated directly with the popular models. This is for example the case for in cup, in bowl, and in box.

Moreover, there are several existing loras on civitai for similar things, and they may suit better your need depending on what you want to achieve (just search in box or in container).

However, for some other concepts, such as in suitcase or in bag, it may be much easier with this lora, unless you are very good at prompting and know how to achieve it directly on your model. Apparently, the effect of the lora on each model is different, so the only way to see whether it makes a difference is to try is yourself.

If you'd like to support me and do more:
https://ko-fi.com/edgonkofi - commissions open!
If you're looking for a >>LoRA Making Tutorial<<

Butterfly Lingerie

Beautiful butterflies...

Your dedicated onlyfans altar, summon your goddess here.

If you love my work, you can buy me a coffee to help me stay awake and cope with life to create more lora and stuffs : ko-fi.com/tauron

My Mix is out : Tauron Hybrid Mix - TauHybrid_Anime_Final | Stable Diffusion Checkpoint | Civitai

DISCLAIMER:

I'm doing this for fun and not for commercial motive, My main purpose is to compose a fictional character using AI with the beauty of all persons above, please support them if you love their works. i do not support any indecent images generated of all persons in this model, I do not support selling generated images of persons in this model. please refrain to what they have been posting online.

All persons in this models are 21+ and not a minor, please DO NOT post explicit nudity, suggestive content, or pornographic content online using this model.

===

Open to suggestion or request

NOTE : I will keep updating more actresses in versions update

===

Usage Guide

--

A. Lora Weight

• 0.1 ~ 0.4 -- is to combine multiple faces or features, or if you experience overbaking

• 0.5 ~ 0.7 -- is a balanced weight to get still accurate feature but with more flexiblity

• 0.8 ~ 1.0 -- is to create accurate representation of the features.

B. CFG Scale

• Lower CFG Scale means AI gets more creative and less overbaking problems.

• Higher CFG Scale means results get very accurate with training sets and more prone to overbaking problems.

• Dynamic CFG Threshold, using this method allows you to bump up quality and achieve photorealistic easily, but with lycoris i found out recently that this method actually lower success rate, so only use when you are composing with a Lora / model that works fine with this method.

C. Sampling Steps, High Res Fix & Upscaling

• I like to use sampling steps at 50, with high res step at 20, or 1/2 of sampling steps.

• I find that 2M Karras works better and faster than SDE Karras. with ERSGAN 4x+ method + ERSGAN 4x for Upscaling. This method actually gives quite good result without being too sharp or too plasticky

• Upscaling - GPFGAN @0.25, Codeformer @0.3 & Codeformer Weight @0.15 seems to yield superior results to me.

--

Combining with Attribute LoRa / LyCoris

You may need to do your own testing, but using character weight 0.2 below attribute lora weight yields superior results, if for some reason one features persists not showing or overbake, you may want to lower the weight of the overbaked feature Lora, and then use emphasis () to boost the specific attribute you want.

DDetailer Script

If subject face is small / far away use: DDetailer Script with bbox anime face with default settings except denoiser at 0.2-0.3. You can combine this LoRa with a lot of other Style / Attire / Pose / Acts LoRa, Also works well with character LyCoris.

How to install this script? I kinda forget how because its a bit complicated beacuse you need to install stuff from venv. but a quick google search will land you somewhere in tutorial page.

Please leave a review if you have any critiques or if you find this LoRa good.

Maria Fianna Albert (Ore dake Haireru Kakushi Dungeon)

All uploaded images are generated by YesMix

options

LoRA weight 0.7~0.8

Sampling steps 30 or Higher

Recommend Hires

Trigger words

• Default

  maria fianna albert, maria, ore dake haireru, bangs, blonde_hair, blue_eyes,breasts, long_hair,cowboy_shot, dress, armor,skirt,thighhighs, white_dress, white_gloves, black_legwear,

This is a model for making eyes a bit better.

You can use it alongside existing models to generate txt2img / img2img or use it with inpainting to fix existing images.

While using alongside other LORA/LyCORIS it's best to not overdo the weight and keep it around 0.2 - 0.6

With inpainting you can use higher weights, remember that you can control the strength by playing with the denoising.

The extraction LoCon of Ice-Lolly model.
雪糕原炼模Ice-Lolly的提取LoCon。
The LoCon can also have better thighhighs and pantyhose but possibly poorer limbs.
该LoCon同样可以有更好的雪糕但是可能肢体较差。
The limbs can be improved to a certain extent using MIDD layered fusion with lower fusion weights.
可以使用MIDD分层融合以及较低融合权重来在一定程度内改善肢体。

this is a certain net meme.

This LoRA is LoCON.

need https://github.com/KohakuBlueleaf/a1111-sd-webui-locon if you want to use it in the web ui.

This LoRA can be combined with Character LoRA.

If you use this with other character LoRA, you need to adjust this LoRA around 0.8-1.3 and character LoRA around 0.8-1.3.

Ribbons and hair ornaments may be attached due to learning data.

Specify a negative prompt if you want.

とあるネットミームです。

LoCON学習なので、automatic1111-webuiで使用する場合は拡張が必要になります。

キャラ学習をしたLoRAとの併用も可能のようです。相性によってはあまりうまくいかないこともあるかもしれません。線や塗りも他のLoRAに引っ張られることがあります。

使用する際は適宜0.8～1.3程度適用する強度を調整する必要がありそうです。

学習データの都合上、リボンや髪飾りが湧き出してくることもあるのでNegative promptに入れるなりして各自対応してください。

Isabell Schneider aka HoneyPuu, is a German streamer on Twitch

Since I don't own a GPU and the only way to continue creating LoRAs is to rent GPUs (free Colab, prohibited Stable Diffusion, and remote UIs), I have chosen to do so with Paperspace. Please check out my Ko-Fi page if you are interested in supporting me.

This is a LoHa LoRA! You need the extension!
Why? LoHa have virtually no effect in output art style and they do a lot better with multi-concepts.

Trigger words for V1.20:

<lora:Hulkv120:0.8>(hulk)

Follow the steps in the Model Version notes to reproduce my images.

Sample images generated from Hassaku, CAN and ProjectAIO.

All negative embeds should be easy to find. and Prompts should be included in the sample images! Click the little icon in the bottom right to see it.

It was trained on 99 512x512 images scraped from the internet and 20 epochs using animefull_final.

Danbooru tags generated by Waifu Diffusion 1.4 Tagger and fine tuned for specific tags.

This LyCORIS has been trained after carefully selecting 96 images. All photos have been enhanced to make the most of them. I hope you like it! Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

akikawa yayoi (umamusume) Lora (LYCORIS - loha) ver.1

※Please do not post NSFW content in places where it can be seen by many people, such as on social media.

I promise! 😁👍

※NSFWはSNSなど、多くの人目につく場所に公開しないようお願いします。

約束だよ😁👍

I'd be happy if you could follow and support me to increase my motivation.

モチベーション向上のためにフォローや支援してくれると嬉しいぞ。

👇😎👇

https://twitter.com/oLAcle1220

https://patreon.com/olacle

Trigger words

• base

  • Trigger word

    • akikawa_yayoi

  • custom prompt (chara)

    • blue eyes,long hair,orange hair,white hair,two-tone hair,multicolored hair,skin fang

    • horse girl,horse tail,horse ears

  • custom prompt (president's clothes)

    • hat,hat flower,blue flower,blue rose,white headwear,animal on head,cat on head,dress,white dress,white skirt,long sleeves,puffy sleeves,jacket,blue jacket,cropped jacket,open jacket,open clothes,ascot,folding fan,socks,blue footwear,buttons,

Recommended settings（おすすめ設定）

weight 0.5~

Please use the "horse girl, horse tail, horse ears" prompt if you want to include ears. If no clothing is specified, the default outfit worn by the chairman will be generated. To use other clothing, please adjust the weight accordingly.

耳を出したい場合などは「horse girl,horse tail,horse ears」のプロンプトを使用してください。

服装を指定しない場合は基本理事長が普段来ている服が出ます。

その他の服を着させたい場合はweight を調整してください。

Based on Animefull-pruned model

LyCORIS by KohakuBlueleaf

Please install extension on WebUI for using LyCoris.

Recommended options

• LoRA weight around 0.7

• Trigger words

  • Base

    • mika \(blue archive\)

  • default clothing :

    • high heels, dress, wrist scrunchie, white dress,capelet, pumps, pantyhose, frilled dress, black footwear, white pantyhose, frills, white capelet

  • hairstyle :

    • single hair bun

    • double bun

    • ponytail

General instruction: Use it. If your model is 2.5d or 3d-leaning(like mine) - please use negative 3d and realistic for more comic-like results.

Seventh Lazy Style Lora.

Based on ~380 image dataset of specific style.

100% SFW.

Generally, this is not supposed to be used for anythin complex, because it was not tagged in a way to be, but this one is actually pretty editable and generalized. I tried new ways of promting to get more objects covered - seem like it worked, i had fun trying different stuff out.

Retro cars are fucking banger if you ask me.

GPTS Lazy Styles will have a lot of bleed, lower than average editability, but i guess starting from this one it'll be decent :D

This one is pretty editable, and for models that are not 2D you could even increase weight to like 1.25 to get more comic more better(or negative 3d/realistic).

Lighter version might be available later.

It's time to take it to the next level.

This LyCORIS incorporates five *-punk genres: steampunk, valvepunk, dieselpunk, teslapunk, cogpunk (other bundles will come later)

The prompt for the preview picture grids is

<lyco:allpunk:.9> { valvepunk | cogpunk | steampunk | dieselpunk | teslapunk | (cogpunk:0.6), (steampunk:0.6) } [concept]

where [concept] is

• coffee machine

• toaster

• battle mech

• spaceship in space, stars

• tank

Keep me motivated and caffeinated:

https://www.buymeacoffee.com/konyconiN

https://ko-fi.com/konyconi7992

Recommended resolution: 512x768

lora weight: 0.8-1.0(I use Additional Networks)

Simpling method: DPM++ 2M Karras

Steps: 20

Model: "Chilloutmix_NiPrunedFp32Fix" or "Chilloutmix"

CFG: 7

No restore face

Hires upscaler: R-ESRGAN 4x+ , Denoising strength: 0.5

Statement

Production of NSFW works is prohibited

The model and its products shall not be used for commercial or profitable exchange

She might look like someone from real life, but she is not what you think. She is a totally fictional character.

!! PLEASE READ !!

If the original artist requests a takedown, I will immediately remove this model off of CivitAI. Thank you. Their links are located here and you can view their work here. Meanwhile, please do not monetize your generations.

Install the LoCon extension here. Or use that extension that's displayed in the blue box.

Resource used: https://civitai.com/models/22530

Weight: 1.0

Trigger: there's none

These sample images do not contain (masterpiece, best quality:1.2) and the two negative embeddings I usually have.

This is a LoHa model trained from 100+ images about lineart girl&animals from 2 different sources. I use multires noise to generate brushwork of dataset.The tone is very close to Longan's false seed peel,and that's where the name longan comes from.

You can use the LoHa to design figures or animals with monochrome,spot color.

Trigger is longan, and brushwork will surprise you

Weight 0.9-1.0 will perform better on scenery&realistic figures,also you can try new way to use,lol

Trained on ~500 Disco Elysium style images (official art + a lot of fan art), mostly portraits (512x768).

Has a lot of variety.

Trained with Anything v5, but is compatible with lots of other models, giving unique results.

Use abstract prompt (maybe at a low weight) to get more artsy if you need to.

You can try using negative prompts (easynegative etc.) but it will result in a different style.

>>> Remember to play around with CFG Scale and LoRA weight <<<

This model turned out very different from my old Disco Elysium hypernetwork. I added a lot more art to the dataset. Please share your results!

My first LoHA/LoRA.

You need the stable diffusion LyCORIS extension in order to make this work. Otherwise you'll get an error about the LoRA layer being named wrong or something along those lines.

Base model is NAI, trained on 228 solo images of angel dust. Works fairly well on the higher end of values (0.6-1.0). 10 epochs, 3 repetitions. 2280 steps in total.

I recommend using this LoHA on a model that's good at generating furry features, such as fluffy-rock, crosskemono, or a merge between the two.

Used a combination of the LyCORIS Experiments and "How to make a LoRA" guide on Civit to figure out how to make it.

I hope you'll enjoy it!

A lora for throatfucking

trained on 97 images

Image info

• Model used abyssorange2hard

• bad hands embedding

• bad artist embedding

• bad prompt embedding

• deep negative embedding

Notes

• just use a similar prompt from the examples

• didnt prune any tags for more freedoom when generating

Training settings?

The most important part of training is getting a good dataset!

Training settings can be found by installing the additional networks extension in auto1111

Extensions -> available -> load from -> additional networks

Restart webui

place lora into stable-diffusion-webui\extensions\sd-webui-additional-networks\models\lora

To view the training settings

Go to additional network tab -> choose model -> training paramters

Manga is different from Anime. Manga style looks like Anime, but it is mainly Lineart. Anime has no line-hatching for shading, manga does.

Manga is different from comic. Manga reads from right to left. Comic reads from left to right. Also manga book often focuses in a story at a certain timeline, not like comic book with many somewhat related stories binding together in various ways.

This model is strong. Suggesting weight 0.4 to 0.7 depending on based model. Higher weight might have panels and speech bubbles.

in V3, the demo manga pages generated -manual speech bubbles n title text- with variation of the prompt

2boys, Harry potter, black_hair_glasses, Draco Malfoy, blond hair, male body focus, sorcerer man dress, (pointed hat), <lyco:mangaV3:.7>manga style lineart monochrome, bw

anypastel, dpm++2m, steps 30, no upscale.

Виталий Громяко, в Интернете более известный как Папич — популярный игровой стример, ставший известным благодаря игре Dota 2. Бывший участник Twitch. Среди геймеров он известен под ником EvilArthas. Является известной личностью среди игрового сообщества и героем мемов Рунета. В Сети у него есть множество фанатов, которые снимают нарезки с его стримов. Известен также множеством собственных оригинальных выражений, породивших своеобразный говор язык рофланов.

This LyCORIS has been trained after carefully selecting 68 images. All photos have been enhanced to make the most of them. I hope you like it! Any feedback would be highly appreciated!

From my experience it runs best with Edge Of Realism, pirsusEpicRealism, epicDiffusion ,RealisticVision, CyberRealistic, Deliberate checkpoints, among others, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

Notice: This free work is intended for art and illustration purposes of emergent technology and not intended nor' trained for making illicit images, as the author I've taken care to not use any private data in the training process or designed it for such purposes. By downloading this model you are agreeing with the aforementioned intent and that you will not use my work to create / share images of a potentially immoral and / or illegal nature.

This art style model I trained had the biggest dataset by far with a total of 927 images between public ones on his Pixiv scraped from Gelbooru and paid ones from his FANBOX leaked on Kemono Party (which include explicit variants). I had to edit quite a lot of images using Lama Cleaner and delete others which deemed untrainable to me, I also edited those explicit ones having censorship with an inpaint model I used in ChaiNNer for bars and WebUI for mosaic (since DeepCreamPy and similar methods like DeepMosaic both suck ballz and are very outdated, wish we had a new model and program in order to demosaic hentai/JAV shit in general besides JAVPlayer, which would hopefully work on Linux too...). Also I included both his old and current style together since I love 'em both, so expect this model to sometimes shift randomly between both styles with each generation

Anyways, this artist right here is another favorite of mine, especially since he's HEAVLY obssessed with pantyhose as much as I do, the way he draws the fabric of them is so astounishing to me. Even the faces he draws are very adorable and he adapts the usage of 3D for his backgrounds which he then traces over lightly, although it may not always generate them in 3D all the time, as it mostly generates simplified 2D ones with not much detail due to the dataset I suppose. But it does one hell of a job at nailing his distinct pantyhose shading very well which I'm thankful of, although this doesn't mean you should ALWAYS generate females wearing pantyhose like he always draws them wearing those on, it also applies it to other kinds of legwears as well such as thighhighs, kneehighs and leggings too I guess (or you could just generate 'em bare legged for all I care, just don't post your generations with this or other style models of mine on Pixiv, since I read that they don't allow images imitating other artists anymore).

So... They have done numerous porn out there, what's left to enjoy?

Well, some reasons maybe that they are retired, or getting old.. or you like younger version of them, or, maybe you just want to put them into a cosplay, a specific play that they never done in porn maybe? or you want to mix them with your other waifus and create some beautiful angel. Now thanks to SD, the possibilities is limited only your imagination and prompt, or maybe bad hands... or some mutated limbs, let us pray to the god of Ai that we are protected against such evil.

If you love my work and want to support me: ko-fi.com/tauron

[EDIT : My model is out, check it: Tauron Hybrid Mix - TauHybrid_Anime_Final | Stable Diffusion Checkpoint | Civitai]

===
DISCLAIMER:

I'm doing this for fun only. my main focus is to generate a fictional character based on the beauty of all persons in this model. I do not support generating indecent imagery more than what they have been doing online, or selling generated image for commercial purpose.

If you love their work please support them, all models are 21+. Please DO NOT post explicit nudity, suggestive or pornographic content with the realistic image of the persons in civitai.

===

Open to suggestion or request

NOTE : I will keep updating more actresses in versions update

===

Usage Guide

Lora Weight = 0.1-0.4 (is best for combining, be mindful that your model may contain bias towards specific face such as korean or asian), 0.5-0.7 (is good to retain original face, while not too constraining your prompt), 0.8-1 ( is to generate accurate features of the original person ).

Use Dynamic CFG Threshold

• Mimic = 10 | Clamp percentile = 99.5

• Min. Mimic Scale = 0, Cosine Up

• CFG Scale Sch. = 15, Half Cosine Down

NOTE : I realized that with LyCoris Person, dynamic CFG actually lower success rate. without dynamic CFG result are more highly successful at the cost of a little less detail.

If subject face is small / far away use: DDetailer Script with bbox anime face with default settings except denoiser at 0.2-0.3. You can combine this LoRa with a lot of other Style / Attire / Pose / Acts LoRa, Also works well with character LyCoris.

Please leave a review if you have any critiques or if you find this LoRa good.

NEW VER:

This version has been retrained and it is recommended to use Perfect World with a weight of 0.6-0.8. I tested some clothes and poses that were not in the data set, and they turned out decently. The next step will be to consider adding more expressions and actions.

该版本重新训练，推荐使用Perfect world，权重0.6-0.8.

测试了一下不在训练集内的衣服和动作，姑且还行。下一步考虑加入更多的表情和动作。

****

Why is it called Almighty, because I hope the Lora can be used with pose and clothing.

The current version: more than 300 pictures, a dozen Concept. recommended with realistic. Can be combined with pose lora, using Basil_mix_fixed can reduce limb collapse. Clothing lora need to increase some weight.

I recommend using Basil_mix_fixed, followed by Perfect world. other models are also available, but I still prefer Basil_mix_fixed, because the most restored. Weighting 0.6-0.9.

为什么叫万能，因为我希望该Lora可以搭配各种动作和衣服使用。

目前版本：300多张图，十几个Concept。推荐搭配realistic使用。可以结合动作lora，使用Basil_mix_fixed可以减少肢体崩坏。衣服lora需要加大一些权重。

推荐使用Basil_mix_fixed,其次是Perfect world。其他模型也可以，但是我还是更喜欢Basil_mix_fixed，因为最还原。权重0.6-0.9。

内置的衣服： Included clothing:

1.Advent_Children_costume,midriff, crop top,skirt, swept bangs, bare shoulders, sleeveless, black skirt, gloves, bangs, collarbone, belt,

2.chinese_dress_purple,thighhighs, double bun, hair bun, chinese clothes, dress, china dress, jewelry,

3.chinese_dress_dragon,double bun, dress, jewelry, hair bun, cleavage cutout, chinese clothes, earrings, gloves, bracelet, cleavage,

4.clothes_lift,nipples, breasts, shirt lift, large breasts,

5.sliver dress,

6.golden dress,

7.Honoka_costume,swimsuit, breasts, bikini, torn clothes, shorts, torn thighhighs,navel, front-tie top, belt, short shorts, front-tie bikini top, bikini top only, jacket, cleavage, open clothes, open jacket,

8.Maid,breasts,maid headdress, cleavage,alternate costume, apron, enmaided, ,jewelry, earrings, leotard, wrist cuffs

9.Standard_clothes,crop top,gloves, suspenders, elbow pads, midriff,  fingerless gloves, red footwear, black thighhighs, black skirt, elbow gloves, suspender skirt, boots,  tank top, navel,sports bra, abs, bare shoulders, zettai ryouiki, pleated skirt, toned, miniskirt

10.nsfw,bondage, bound, arms behind back, nude, nipple rings, rope, pubic hair