Shampoo, one of the favorite characters of Ranma 1/2

english↓

This is lora by kitakoji hisui, a member of the virtual youtuber group nijisanji.

The strength of lora is recommended to be around 0.8~1.

Trigger words for the student version:1 girl,hsi,short hair,school uniform

Trigger words for Adult version:1 girl,hsi,long hair,otft2

It also works well if you put the words that occur frequently by putting lora itself through analysis.

This may not be a lora with satisfactory accuracy since it was made as a trial.

Please enjoy it if you like.

https://www.anycolor.co.jp/guidelines/

japanese↓

これは、バーチャルyoutuberグループのにじさんじに所属している北小路ヒスイのloraです。

loraの強度は0.8~1くらいがおすすめです。

学生バージョンのトリガーワード:1 girl,hsi,short hair,school uniform

大人バージョンのトリガーワード:1 girl,hsi,long hair,otft2

また、lora自体を解析にかけて頻出するワードを入れてみると効果があります。

お試しで作ったので、満足のいく精度のloraではないかもしれません。

よければお楽しみください。

Who? : She is Catura's mother in Granblue Fantasy. Name is unknown.

Weight :0.7~1.0

Trigger :caturamama, closed eyes, glasses,

Official Costumes Trigger : ushikimono

Issue:

• Ears and horns, cow ribbons(?) are very unstable.

• If used another outfit, the official costume may be affected.

Notice:

There is no source for eye color, if you want the same as Catura, "yellow eyes" would be good.

Never take off her glasses.

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

trained on 768 by 768

I trained a LoCon on 200 images of SakimiChan and got this.

My first uploaded model. Used this model for samples. Recommended weight 0.75 (depends on your checkpoint though).

This LoCon tries to capture the artstyle of the mid-episode eye catchers from the anime Gurren Lagann.

Use weights <1 .

Unfortunately, almost all images include the anime logo, so use "copyright name" in the negative prompt if it shows up.

Welcome to Insta Thot Series Vol. 2! Realize your thot fantasy 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. It take less than 5 mins to like or leave a review while it takes me hours or even days to make Lora. Liking / Reviewing helps the thread alive.

Check vol.1 --> Tau's Insta Thot Vol. 1 [Fin] - 01. [Req] @Imazhai | Stable Diffusion LyCORIS | Civitai

Best to use with my mix: 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. Refer to CivitAi TOS for more details.

===

Request or Suggestion Rule:

1. No request of minor!

2. The requested should have public data (no private)

3. Should have more than 30 good quality images

4. Since all requests or suggestion are free, I will still need to like your request, otherwise its a work not a hobby for me.

5. Be a decent human being and leave a review if your requests is up It takes you only 5 mins while it takes me hours, electricity, and efforts to do your requests.

6. If you don't get what you want, do not review bomb, there could be other creators models or lora that can get better results than mine, use theirs instead, and leave a good review on them too, support any other creators for their hardworks!

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 at 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 / ADetailer 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.

Enjoy.

Someone asked me for Lieutenant Dan after having pubished a Tom Hanks model based mainly on Forrest Gump. So here it is ! :)

This LyCORIS has been trained after carefully selecting 28 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.

Please check out my Ko-Fi page if you are interested in supporting me.

LoRA commissions open! -> Fiverr

This is a LoHa LoRA! You need the extension! Why? They do a lot better with multi-concepts.

This is retrained from mild's dataset! He was kind enough to share his dataset.

Trigger words for V1.10:

positive: <lora:blackSclera110:1>(black sclera)

I suggest using adetailer for far away shots... Just remember to call the lora and "black sclera" on it.

You can also try to prompt for an iris color like "blue iris" or "green iris" but theres a high possibility of it not generating the expected color.

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

Sample images generated with a merge of Nothingv2.3 + my Clay Mann Style Lora.

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 46 512x512 images provided by mild, 20 epochs using Animefull-final-pruned for the Training Model.

Danbooru tags generated by Waifu Diffusion 1.4 Tagger.

Twitter:https://twitter.com/sudekorasa60425

Sampler: Euler
Steps: 60
Resolution: 512x768
CFG Scale: 7

Hires Fix is recommended.
SwinIR_4x
Denoising Strength 0.1
Upscale by 2

Commercial use of LoRA is strictly prohibited.
Furthermore, please be solely responsible for any image generation using LoRA.

当LoRAの商用利用での利用は厳禁です。
また、当LoRAを使用しての画像生成は全て自己責任で行なってください。

Recommended resolution: 512x768

lora weight: 0.8

Simpling method: DPM++ 2M Karras

Steps: 20-35

CFG: 7-8

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

This LyCORIS has been trained after carefully selecting 65 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.

Micha Jawkan is a character of Senki Zesshou Symphogear GX.

Recommended weight : 0.7~1 (0.8 might be best.)

All uploaded images were generated by anything-v4.5.

The VAE used is kl-f8-anime2.ckpt.

Also, EasyNegative is used.

If the costume generated is not good, some prompts of the following may help.

• frilled capelet, pink shirt, mechanical hands, claws, frilled cuffs, mechanical legs, shoes

Enjoy!

【日本語版】

ミカ・ジャウカーンは戦姫絶唱シンフォギアGXに登場するキャラクターです。

おすすめweight：0.7～1 (0.8が一番いいかも)

アップした全ての画像はanything-v4.5で作成しました。

使用したVAEはkl-f8-anime2.ckptです。

また、EasyNegativeも使われています。

もし生成される衣装が良くない場合、以下のプロンプトの中に役立つものがあるかもしれません。

• frilled capelet, pink shirt, mechanical hands, claws, frilled cuffs, mechanical legs, shoes

楽しんで！

Support me on buymeacoffee if you like my work.

Göksel is one of my all time favorite singer that I listened a few songs from my native language. So she is very special for that reason.

Recommended settings : 0.7-0.9 weight with hiresfix 0.5 with dpm++ 2m karras sampler.Hope you like it ^^

Credits (Because civitai doesn't recognize them even though I installed the extension):

EnvyBetterHands LoCon : https://civitai.com/models/47085?modelVersionId=55199

Sharpness Tweaker LoRA : https://civitai.com/models/69267/sharpness-tweaker-lora-lora

Simple pointed toe stiletto heels without ankle strap (无带尖头高跟鞋) : https://civitai.com/models/81600/simple-pointed-toe-stiletto-heels-without-ankle-strap

Trained with 1.5 pruned on 1024px

Recommended: VAE(84000...)

All images were created without hires fix or face restoration.

x光 midjourney风格lora，权重设置在0.5-1.3之间，底模选取realisticvision

关键词：xguang，illustration，

可以产出，人物，风景，动物，汽车等等

在此做出预告，我会将细分动作全部在本概念下更新，每一种姿势对应一个版本，请按需切换下载

Here is an update notice. I will update all subdivided actions under this lora. Each pose corresponds to a version. Please switch and download as needed.

版本之间不存在迭代差异，仅有姿势区别

There are no iteration differences between versions, only pose differences

目前已经更新第4个

The 4th one has been updated

好的手需要运气这件事我想大家应该不用我再强调了吧

Good hands need luck. I don’t think you need me to emphasize it.

下面是脱口秀时间，因为比较长还是一个梗，我就不翻译了

The following is the time for the talk show, as it is still a meme, I will not translate it anymore

总结一下就是，目前我试了几个已有的Pov missionary都是简单动作概念，我本来打算将Pov missionary的多细分概念合一，然而失败多次，决定拆分制作

The general content is that at present, I have tried several existing Pov missionary loras, all of which are simple action concepts.
I originally planned to combine the multiple subdivision concepts of Pov missionary into one, but failed many times, so I decided to split it.

我也想一个lora概念全包

是不是很大胆

一个lora概念全包

在Pov missionary这一个

就这个lora

就完全是这个lora   分概念全包（指包括什么leg grab、holding another's wrist、torso grab等等）

我最早收集数据集的时候

跟所有群友说的都是一个全包

难以想象

喔所有群友啊 

他们所有人都很兴奋（其实只有我）

一个全包一直说的是一个全包

我一直在做一个lora概念全包的准备

所以当时跟显卡说的时候

我只要你20天

或者只要你25天

10epoch到15epoch

显卡说为什么

我说很简单

我每天炼两遍

一锅lora两个小时

我每天炼两遍

上班炼一遍

下班炼一遍

我每天炼两遍

我炼十天

400多个lora我选一个

一个lora概念全包啊

这个Pov missionary就完了

我选一遍

那十天就够了

然后前面的十天打标

严格的打标

我说只要你10天咱就炼一超全的lora

就挑战性太大了

一点都不能错呀

但是后来我还是放弃了

你知道为什么

我研究了大量的包含复数概念的lora

也有一些大概念lora

还有尤其是最近这2个月

有一些实验性的多概念多角度lora

我自己做的别人做的我都看过

我专门   看研究

也就一直我就一直下不了决心

其实在sex from behind的时候

我就想做一个多概念多角度合一的

到底后来也没有做

你知道最大问题是什么

当然这是一个专业问题

简单的说最大的问题就是，同步拟合

就是同步拟合

因为多个概念啊，拟合没有办法

对于一个多细节的

对于一个多变化的

一个复数概念的一个lora来说

那么适合的一个拟合程度达不到，同步拟合达不到

拟合度达不到

因为你不能正面是这样子

不可能

你从正面到侧面、从握手到抓胸

这一出图就寄了

第一是拟合达不到

无论怎么样

我想来想去

因为我算个内行吧（笑）

我不炼我就知道拟合达不到（其实炼了）

同步拟合实现不了

第二个   作者的精神状态会损失一半以上

作者的精神会损失一半以上

作者的大脑会损失一半以上的san

是看不看不到的

lora就不好做了

lora就不好做了

最后我差不多半个多月的思想斗争

那是个重大决定

一个个训练集都是按照一个lora概念全包准备的

还ps、打标呢

后来说我放弃

 我自己决定放弃

一个人做的决定

一个人的思考

半个多月以后

我放弃以后

就觉得好像就没劲了

不能一个lora概念全包

就觉得好像就不想做了（哈哈哈）

开玩笑啊

就所有人都沮丧而失落（就我自己）

我也很难过

我后来还是放弃了吧

所以制作概念lora是一个学习

它对于一个追求高质量的一个创作者来说

它是破坏性的

一个lora概念全包又能怎么样

你丢掉了质量

丢掉了牺牲了出图成功率

孰重孰轻啊（点头）

Fantasy (and sci-fi characters) in a colorful, painterly style. This is a strong one, so use it at around 0.7 for nice results.

If you like the Sexy Sorceress trope (and who doesn't, really?), this is the LoCon for you. Trigger words are completely optional if you want to focus on a specific type of character, but it'll still make things look cool without a trigger word.

I like this one best around 0.8, and I'm getting the best results by keeping the prompt relatively short and simple and letting it figure out the costumes on its own.

A last second LoHa model. Reminder that the request form is closed.

Your characters may have insufficient training data! Click here.

Weight: 1.0

Trigger + Additional Tags: u.a. school uniform, grey jacket, red necktie, green <skirt / pants>

Doesn't particular do well with green pants though.

The results of your generations are dependent on the additional LoRAs, weights, and models you use, so it may not work or come out as consistent as my sample images. Other reasons may include me using xformers, token merging, and a modified DPM++ 2M Karras sampler. You'll also notice that my prompt includes <lyco:. Note that all my LoRA files are located within my LyCORIS folder. If the model is labeled as a LoRA, it's a simple LoRA. If it's specified as a LyCORIS, it's a LoHa.

Medina Orthinanos from Sword Art Online Alicization Lycoris, trained as a LyCORIS network. Sorry for the bad pun.

Requires extension for use:

https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris

Recommended tags are:

dsmedina, red hair, short hair, green eyes, medium breasts, cape, cleavage, gloves, thighhigh boots

Check out my HuggingFace repository:

https://huggingface.co/Doctor-Shotgun/ds-LoRA

Check out my Pixiv:

https://www.pixiv.net/en/users/90447503

Because it is based on LoRA but is trained with Conv2D

Main tag : Exhibitionism,
Secondary tag : public indecency, public nudity,

Conv2D requires need extensions
https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris
or
https://github.com/kohya-ss/sd-webui-additional-networks

InfiniteMoon/00loha · Hugging Face

经由本人同意制作的lycoris模型，没有触发词，擅长男福瑞，也能画美少女，女福瑞和普通男性效果一般。推荐使用14号模型。非常不建议用V1.2的模型，那个过拟合严重。

di.FFUSION.ai-tXe-FXAA

Trained on "121361" images.

Enhance your model's quality and sharpness using your own pre-trained Unet.

The text encoder (without UNET) is wrapped in LyCORIS. Optimizer: torch.optim.adamw.AdamW(weight_decay=0.01, betas=(0.9, 0.99))

Network dimension/rank: 768.0 Alpha: 768.0 Module: lycoris.kohya {'conv_dim': '256', 'conv_alpha': '256', 'algo': 'loha'}

Large size due to Lyco CONV 256

For a1111
Install https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris

Download di.FFUSION.ai-tXe-FXAA to /models/Lycoris

Option1:

Insert <lyco:di.FFUSION.ai-tXe-FXAA:1.0> to prompt
No need to split Unet and Text Enc as its only TX encoder there.

You can go up to 2x weights

Option2: If you need it always ON (ex run a batch from txt file) then you can go to settings / Quicksettings list

add sd_lyco

restart and you should have a drop-down now 🤟 🥃

More info:

"ss_text_encoder_lr": "1e-07",

"ss_keep_tokens": "3",

"ss_network_args": {

"conv_dim": "256",

"conv_alpha": "256",

"algo": "loha"

},

"img_count": 121361

}

"ss_total_batch_size": "100",

"ss_network_dim": "768",

"ss_max_bucket_reso": "1024",

"ss_network_alpha": "768.0",

"ss_steps": "2444",

"sshs_legacy_hash": "539b2745",

"ss_batch_size_per_device": "20",

"ss_max_train_steps": "2444",

"ss_network_module": "lycoris.kohya",

This is a heavy experimental version we used to test even with sloppy captions (quick WD tags and terrible clip), yet the results were satisfying.

Note: This is not the text encoder used in the official FFUSION AI model.

这几天一直更新同一个姿势有点无聊，穿插一点别的

It’s a bit boring to keep updating the same pose these days,so i made this

可能出现手部错误和肢体错误

Hand errors and limb errors are possible

直立姿势表现不是太好，你最好加上spread legs

The standing posture is not as good as the sitting posture, so you'd better add “spread legs”

可选的tag:

Optional tag:

hand in panties

简单的示例prompt：

Simple example prompt:

1girl,1boy,fingering from behind,grabbing,spread legs,hand in panties

👑 SteelHeart Silverhawks LyCORIS👑

🔧Model base: sd1.4

_____________

📝🌟 Trigger: (SteelHeartQuiron character), <lyco:quiron_SteelHeart_V3Lycoris:1.0>, (silver metalic short hair), metalic futuristic bodysuit, science fiction,, 1980s \(style\), retro artstyle,

_____________

📝 Variation: <lyco:quiron_SteelHeart_V3Lycoris:Y> Y=0.8 to 1.0

📝 Denoising strength: 0.30 to 0.6

_________

🗒️💡Sample:

(masterpiece, best quality, high resolution), ( 1human arm:1.3), (cyborg:0.8), intricate bodysuit, skintight silver armor, (cartoon:1.3),(SteelHeartQuiron character), 1girl, solo, muscle legs, abs, (silver metalic short hair:1.5), metalic futuristic bodysuit, science fiction, clenched hand, a close up of a cartoon of a woman in a space suit, as a retro futuristic heroine, art deco of a space woman, inspired by Rob Liefeld, scifi woman, as a retrofuturistic heroine, 60's sci-fi pinup style, cyborg goddess in cosmos, beautiful cyborg girl pinup, <lyco:quiron_SteelHeart_V3Lycoris:1.0>

___________

🚀 This LORA likes: (best quality:1.1), (masterpiece:1.2), (realistic:1.2), (detailed:1.1),

👇 Negative Prompt: (worst quality:1.2), (low quality:1.1), (normal quality:1.1), lowres,

💡Sample steps: 20 to 39

💡CFG Scale: 5.8 to 10.5

Hire. fix: ok

👇Restore faces: Yes /No

____________

☕ If you feel like supporting my work, please consider visiting my Beer / BuyMeaCoffe ( https://www.buymeacoffee.com/quiron ) ☕

🌟☕If you like my work please support ( https://www.patreon.com/VisionsofAi )☕🌟

As usual, the model is trained with clip skip 1 on top of ACertainty and locon is extracted from the trained model

The locon should work on most models as I show in the xy plot

All the intermediate checkpoints can be found on https://huggingface.co/alea31415/shokei-shojo

It includes the following 7 characters from the series

• Ashuna

• Flare

• ManonLibelle

• Menou

• Momo

• Pandemonium

• TokitoAkari

For Menou you have both

• Menou

• Menou child

For Pandemonium you have

• Pandemonium (the classic one)

• Maya Pandemonium (hair down version, the one after the end of the 5th tome of the novel)

• Centipede Monster (monster version, the one at the 12th episode of the anime)

Maya Pandemonium is only trained on 6 images from the light novel: 2 colorful and 4 monochrome, but it should still be usable if you adjust things properly

The preview of Centipede Monster is only included for the FullCkpt version, because I guess it may make some uncomfortable (and anyway I do not think anyone would ever use it)

I split the styles into 3+1 during training

• aniscreen (3928 images)

• fanart (95 images)

• colorful light novel / light novel (153 images)

As usual I use a lot of regularization images during training, in particular I also include most of my dataset for the in container loha (https://civitai.com/models/66021) here so I guess these concepts are more or less learned as well

The example images are either generated by the LoCon on top of VBP2.3

Or using a certain merge of the trained model

None of the images are generated in one shot

They generally include a lot of random draws, img2img, inpainting, and sometimes use of controlnet and regional prompter (for the case of multiple characters)

--

Just to complain, I really have a hard time figuring out what kind of images violate civitai TOS

I wanted to upload an image with Pandemonium on top of ManonLibelle who is lying in the blood (this one), but apparently this is detected as violation of TOS and gets removed automatically

In the mean time, we can easily find a lot of images that are on civitai that are more violent/horrible/bloody or whatsoever

Otherwise, although probably few people would ever be interested in a model for this

Shokei shojo is basically my favorite light novel, so I kind of feel that I must do it

Btw I saw an image of Menou from this discussion on outpainting of controlnet https://github.com/Mikubill/sd-webui-controlnet/discussions/1464 so I told myself it was time to get this model done

This LyCORIS has been trained after carefully selecting 45 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.

DISCLAIMER: I am NOT affiliated with Majalis in any way. I just liked the game, scraped a bunch of images from the internet and created this model.

This is a style/characters from the game Tales of Androgyny by Majalis duo.

The version v2.0 surprised even myself by how good it got at generating all of the characters. Still struggles with full sex scenes, so there's room for improvement, but single characters are really good.

You should ignore v1.x as it's far worse, leaving it for now, but will probably remove in the future.

Please share your images here, as I would love to see what people can create with this.

The model was trained mostly on NSFW images (it's a pron game, duh!), but it's capable of generating SFW images of characters as well.

It was trained on AnyLora, but it turned out to be pretty portable to other models, so far I've successfully tried FantasticChix-HR and Perfect World.

You should use CLIP Skip 2 (that's what AnyLora uses). You should start the prompt with "majalis" followed by the character you want like "kylira". For optimal results, use "1girl" + "futanari" for all the futa characters (Alma, Urka (orc), goblin, Merrybelle, Quetzal, demonking, darkknight, Dullahan, and much more...) and "1boy" + "femboy" for the boys (Hiro, Kylira, Trudy). Model strength in range 0.5-0.7 works well, depending on how realistic the checkpoint you're using is, more realistic - less model strength.

You can find all of the tags used (and their frequencies) in the metadata of the model.

It's capable of generating penis instead of vagina for futas and traps in sexual positions even without other LoRAs.

This LyCORIS has been trained after carefully selecting 59 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.

LyCORIS of Eva Elfie. Trained at 768x768 with 88 images, a mix of close-ups, waist up and full body shots.

Use eva_elfie as the trigger word.

Sample images were generated with realisticVisionV20, with 0.65 strength then inpainted at 1.0 strength.

This is a LoHa learned in Replicant-V3.0.

To use it, you need a model of Stable Diffusion V2.1-768 series and LyCORIS installed.

(When run on SD1 series models, the following error occurs.

RuntimeError: The size of tensor a (768) must match the size of tensor b (1024) at non-singleton dimension 1).

The file should be placed under the models/LyCORIS folder.

This is created to output a male doll with a gothic feel.

The effect is too strong and may be best used at -0.5 to 0.5.

As a side effect of this LoHa, applying a positive value will increase the realism, and applying a negative value will make more animation style.

The training process uses about 300 images generated by niji journey and about 100 images generated by Replicant v3.

The license is taken over from Replicant v3.

license : https://freedevproject.org/faipl-1.0-sd/

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

これはReplicant-V3.0 で学習した‎LoHaです。

使用にはStable Diffusion V2.1-768 系のモデルとLyCORISのインストールが必要です。

(SD1系のモデルで実行すると、以下のようなエラーが出ます。

RuntimeError: テンソル a (768) のサイズは、非シングルトン次元 1 でのテンソル b (1024) のサイズと一致する必要があります。)

ファイルは models\LyCORIS フォルダの配下に配置してください。

ゴシックな雰囲気の男性ドールを出力するLoHaを目指しました。

効果が強いので、-0.5～0.5程度で使用するのが良いかもしれません。

このLoHaは副次効果として、プラス適用するとリアルさが増し、マイナス適用するとアニメ調になります。

学習過程では、niji journeyで生成した画像約300枚と、Replicant v3で生成した画像約100枚を使用しています。

ライセンスはReplicant v3に準じます。

ライセンス: https://freedevproject.org/faipl-1.0-sd/

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

trained on 768 by 768

I trained a LoCon on 200 images of Nanoless and got this.

LyCORIS of Weronika "Wersow" Sowa

LyCORIS of Tricia Helfer

LyCORIS of Sarah Michelle Gellar

LyCORIS of Toni Collette

LyCORIS of Robin Wright

LyCORIS of Rhea Seehorn

LyCORIS of Olga Skabeyeva

LyCORIS of Priyanka Chopra

It's an experimental LoHa model trained with noise.

This model can achieve more detailed dark tone city&wild background,maybe there's a surprise in terms of composition.

This name is a description of tone of color.

weight 1.0 maybe generates better imgs.

Add no humans, moon, sky, night, scenery to improve better.

Weight 0.9-1.0 will perform better

Not Queens. Not Vay Hek.

For the... huh... For my brothers.

This LyCORIS has been trained after carefully selecting 34 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.

6/3 新バージョン追加

教師画像を一新し、次元数を上げました

以前のバージョンと比較して、奇妙な形で生成される頻度が少なくなりましたが、

代わりに写実的になりました

ドアが抜け落ちる事はもうありません！

6/3 New version added

The teacher image has been redesigned and its dimensionality increased.

Compared to the previous version, they are now generated less often in an odd way, but are instead more realistic.

No more doors falling out!

トリガーワードではないですがthree-wheelerもプロンプトに加えないと、4輪になってしまう確率が非常に高いです

あなたがこのLohaを利用した画像もギャラリーにアップロードしてみてください

追記:ネガティブプロンプトにscooter等を入れると、車になる確率が上がるかもしれません

Sentences from here down are machine translated

It's not a trigger word, but if you don't add three-wheeler to the prompt, there is a very high probability that you will end up with four wheels.

You can also upload images of your use of this Loha to the gallery!

PS: Putting scooter, etc. in the negative prompt may increase the probability of a car!

☕If you like my work, please feel free to donate! https://ko-fi.com/notekaga

[English]

Recommended weight: 0.8 - 1.0

Character Trigger

• hinatsuki mikan

Outfit Trigger

• School uniform(full body):

  full body, school uniform, red bow, white shirt, collared shirt, short sleeves, sweater vest, pleated skirt, loose socks, loafers

• Magical girl(full body):

  full body, magical girl dress, pantyhose, boots, white footwear

• For other outfits, see the gallery prompts.

Negative Prompt

• If the character's hairstyle is not accurate, please enter the following words in the Negative Prompt:

  double bun

[日本語]

推奨ウェイト: 0.8 - 1.0

キャラクタートリガー

• hinatsuki mikan

服装

• 制服(全身):

  full body, school uniform, red bow, white shirt, collared shirt, short sleeves, sweater vest, pleated skirt, loose socks, loafers

• 魔法少女(全身):

  full body, magical girl dress, pantyhose, boots, white footwear

• 他の服装については、ギャラリーのプロンプトを確認してください。

Negative Prompt

• 髪型の再限度が低い場合は、次のワードをNegative Promptに入力してください:

  double bun

english↓

This is lora by makaino ririmu, a member of the virtual youtuber group nijisanji.

The strength of lora is recommended to be around 0.8~1.

Trigger words:1 girl,rrm

It also works well if you put the words that occur frequently by putting lora itself through analysis.

This may not be a lora with satisfactory accuracy since it was made as a trial.

Please enjoy it if you like.

https://www.anycolor.co.jp/guidelines/

japanese↓

これは、バーチャルyoutuberグループのにじさんじに所属している魔界ノりりむのloraです。

loraの強度は0.8~1くらいがおすすめです。

トリガーワード:1 girl,rrm

また、lora自体を解析にかけて頻出するワードを入れてみると効果があります。

お試しで作ったので、満足のいく精度のloraではないかもしれません。

よければお楽しみください。

Look at the examples on the metadata of each image, you can trigger the Suite with [ bodysuit ] [ uniform ] or [ military ] combined.

This is a test to see how AI transforms a male character into a female character.

 It originated from a whimsical idea, wondering what would happen if intentionally misleading instructions were given to AI.

 Since the gender characteristics in 2D anime-style images are not as distinct, the training dataset primarily consists of realistic images.

 Depending on the specific Lora model used, weight  from 0.2 to 0.5 can produce noticeable gender transformations. However, if the weight exceeds 0.7, it may lead to overfitting.

这是一个测试，让ai把男的当成女的。

源于一个脑洞，如果你故意教ai错误的东西，会怎么样？

因为二次元图像的性征不明显，所以训练集多采用了三次元的，所以整体的风格偏三次元。

根据不同的Lora，权重从0.2到0.5能看出性别的明显变化，如果大于0.7的话，容易过拟合。

LoHa of Little Cocon (umamusume). Please install extension, a1111-sd-webui-lycoris, on WebUI. LyCORIS, not LoCon

After installing extension, put LoHa file in your stable-diffusion-webui/models/LyCORIS folder and write LyCORIS notation in prompt to applying LoHa. <lyco:your_loha_name:weight>

All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍

NOTICE

• After careful consideration, I decided not to produce LoRA of the characters below.

  • Otonashi Etsuko, Akasaka Misato, Hosoe Junko

• If anyone wants the characters above, I will consider making them.

• Except for the above, I will make the rest of the characters in Umamusume: Pretty Derby, unless there is a special notice.

Recommended options

• LoRA weight around 0.7

• Trigger words

  • Base

    • little cocon \(umamusume\)

  • Original Art :

    • hair bow, striped bow, hood down, black hoodie, black choker, black collar, frilled sailor collar, drawstring, puffy long sleeves, sleeves past wrists, blue dress, chest harness, o-ring, short dress, frills, kneehighs, knee boots, striped socks, blue footwear, fishnets, asymmetrical legwear, mismatched legwear

  • Tracen summer uniform

    • tracen school uniform, summer uniform, serafuku, puffy short sleeves, purple bowtie, horseshoe ornament, sailor collar, sailor shirt, purple shirt, white skirt, pleated skirt, frilled skirt, miniskirt, zettai ryouiki, white thighhighs, brown footwear, loafers

  • Tracen winter uniform

    • tracen school uniform, ear ornament, hairclip, winter uniform, puffy long sleeves, sailor collar, purple shirt, white bowtie, horseshoe ornament, purple skirt, pleated skirt, purple thighhighs, brown footwear, loafers

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

  • DPM++ Series(SDE Karras, 2M Karras, etc.)

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

      • Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other Latent series.

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 69 images consist of

  • 4x low quality 21 images

  • 8x medium quality 29 images

  • 16x high quality 19 images

  • => 1 epoch = 620 images

• replaced character feature tags with little cocon \(umamusume\)

  • 1girl, horse girl, horse ears, simple background, swept bangs, blonde hair, green hair, streaked hair, horse tail, short hair, animal ears, solo, orange hair, bangs, multicolored hair, brown hair, hair between eyes, blue hair, green eyes, two-tone hair

  • I think these tags are the essentials of Little Cocon or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training data or metadata of LoRA file.

All uploaded images are generated by AOM3A1

And, special thanks to my best friend. DH.

This LoRA is not official or authorized.

I check Twitter every day, so I will respond when I receive a DM.
ツイッターは毎日チェックしているのでDMが来たら対応致します。

LoHa of Bitter Glace (umamusume). Please install extension, a1111-sd-webui-lycoris, on WebUI. LyCORIS, not LoCon

After installing extension, put LoHa file in your stable-diffusion-webui/models/LyCORIS folder and write LyCORIS notation in prompt to applying LoHa. <lyco:your_loha_name:weight>

All of my works is free. If you'd like to support me, feel free to buy me a coffee. 👍

I still confuse... What is right her name? Bitter Glasse or Bitter Glace? The browser point out a typo(glasse), so I made it glace.

NOTICE

• After careful consideration, I decided not to produce LoRA of the characters below.

  • Otonashi Etsuko, Akasaka Misato, Hosoe Junko

• If anyone wants the characters above, I will consider making them.

• Except for the above, I will make the rest of the characters in Umamusume: Pretty Derby, unless there is a special notice.

Recommended options

• LoRA weight around 0.7

• Trigger words

  • Base

    • bitter glace \(umamusume\)

  • Original Art :

    • ear ornament, arm strap, black choker, fingerless gloves, red gloves, white shirt, crop top, hooded jacket, hood down, sleeveless jacket, red jacket, cropped jacket, open clothes, brown belt, red shorts, short shorts, bandaid on leg, kneehighs, black socks, black footwear, boots

  • Tracen summer uniform

    • tracen school uniform, summer uniform, serafuku, puffy short sleeves, purple bowtie, horseshoe ornament, sailor collar, sailor shirt, purple shirt, white skirt, pleated skirt, frilled skirt, miniskirt, zettai ryouiki, white thighhighs, brown footwear, loafers

  • Tracen winter uniform

    • tracen school uniform, ear ornament, hairclip, winter uniform, puffy long sleeves, sailor collar, purple shirt, white bowtie, horseshoe ornament, purple skirt, pleated skirt, purple thighhighs, brown footwear, loafers

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

  • DPM++ Series(SDE Karras, 2M Karras, etc.)

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

      • Latent, Latent (nearest-exact), Latent (bicubic antialiased) or other Latent series.

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 26 images consist of

  • 4x low quality 10 images

  • 8x medium quality 9 images

  • 16x high quality 7 images

  • => 1 epoch = 224 images

• replaced character feature tags with bitter glace \(umamusume\)

  • 1girl, [a-zA-Z\s\-]+hair[a-zA-Z\s\-]*, horse[a-zA-Z\s\-]+, [a-zA-Z\s\-]+eyes[a-zA-Z\s\-]*, [a-zA-Z\s\-]+ears[a-zA-Z\s\-]*, [a-zA-Z\s\-]*bangs, [a-zA-Z\s\-]+\\\(umamusume\\\)

  • I think these tags are the essentials of Bitter Glace or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

• other training settings is included in training data or metadata of LoRA file.

All uploaded images are generated by AOM3A1

And, special thanks to my best friend. DH.

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

trained on 768 by 768

I trained a LoCon on 200 images of Ginhaha and got this.

Information

Created for another lora break. It's okay if the face is output.

他のloraの息抜きに作成されました。顔が出力されればまあいいかのノリ。

weight: 0.8

trigger: satsuki

base: bangs, blush, brown hair, large breast, purple eyes, short hair

school uniform: bow, school uniform, serahuku, short sleeve, plated skirt

casual: belt, blue shirt, denim shorts, long sleeve

school swimsuit: school swimsuit

ーーーーーーーーーーーーーーーーーーーーーーーーーーーーーーー

Trigger words are required, others are optional.

トリガーワードは必須です、他はお好みでどうぞ。

Technically a failed attempt in that I wanted to be able to prompt for different types of bulges but couldn't figure out a way to make it work. So instead this model spits out a random kind of bulge. There are ~4 types

1. Sideways bulge

2. Upwards pointing bulge

3. Downward pointing or along leg

4. Ball bulge(Technically not a penis outline but eh)

Basic Prompt:

• 1girl, solo, futanari, penis_outline

There are terms that were intended to trigger the first three bulges but they didn't seem to work.

1. sidebulge

2. upbulge

3. downbulge

I'm open to tips on how to better train different concepts in the same model, it's very possible I just didn't use enough images.

銃座は上手く再現されません

The seat of the gun is not well reproduced.

Something I haven't seen anyone post so here's my attempt. Seems to work reliably enough for what I tested with. Works best with flaccid members and standing subjects in my experience.

Basic Prompt:

• 1girl, futanari, futapee, penis,<lora:FutaPee-V1:1.0>

!! PLEASE READ !!

If the original artist requests a takedown, I will immediately remove this model off of CivitAI. Thank you. Please support their work here. Meanwhile, please do not monetize your generations.

This is a requested style. Reminder that the request form is closed.

Weight: 1.0

Trigger: there's none

The results of your generations are dependent on the additional LoRAs, weights, and models you use, so it may not work or come out as consistent as my sample images. Other reasons may include me using xformers, token merging, and a modified DPM++ 2M Karras sampler.

This model was created in cooperation with two of my friends MalcolmRey and TheBiggestSub, special thanks to them :)

LyCORIS of adult model Zafira. Trained at 768x768 with 70 images, a mix of close-ups, waist up and full body shots.

Use zafira as the trigger.

Sample images were generated with realisticVisionV20, with 0.65 strength then inpainted at 1.0 strength.

I'm testing out the Dadaptation optimizer as outlined in this guide: https://rentry.co/dadaptguide

The results seem pretty good, though I'm definitely running into some problems from my inexperience. I think I included too many NSFW images in the training set, since any prompting outside of her two main outfits will more often than not leave her topless. Eye color also needs to be specified since that wasn't consistent throughout the training data either. Still working on how many steps looks accurate without overfitting. I'm including two versions so people can use them as they like.

Trained for two main outfits:

Regular:

zooey_reg, armor, gloves, thighhighs, breastplate, blue skirt, bare shoulders

Summer:

zooey_summer, swimsuit, hair flower, bikini, frilled bikini

General helpful tags:

tan, dark skin, dark-skinned female, red eyes, long hair, hair between eyes

So far it favors her summer look, so add "head flower" to negatives to remove it.

This LyCORIS has been trained after carefully selecting 54 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 is a requested character. Reminder that the request form is closed.

Your characters may have insufficient training data! Click here.

Weight: 1.0

Appearance: symbol-shaped pupils, +_+

Default Outfit: lemalindef, double bun, short dress, clothing cutout, underboob cutout, ribbon, white pantyhose, thighband pantyhose

Muse: lemalinmuse, double bun, frills, white dress, tutu, pantyhose

Sleepy Sunday: lemalincasual, double bun, criss-cross halterneck, white dress, choker

Mercredi at the Secret Base: lemalinmerc, double bun, headphones around neck, bikini, thigh strap, arm garter, scrunchie

Listless Lapin: lemalinlapin, twin braids, fake animal ears, playboy bunny, leotard, wrist cuffs, highleg, white pantyhose

Add white hair and blue eyes if necessary.

The results of your generations are dependent on the additional LoRAs, weights, and models you use, so it may not work or come out as consistent as my sample images. Other reasons may include me using xformers, token merging, and a modified DPM++ 2M Karras sampler.

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

I trained a LoCon on 200 images of After_sex images and got this.

Pretty girls in sexy clothes with amazing hair.

No trigger words are necessary, but try things like "beautiful bombshell hair" and "outlandish bombshell hair" for some extra fun.

Between releases of my fully trained checkpoints I'll be trying to release experiments like this for upcoming concepts.

Trained on ~150 fully captioned images. I have tested it on various checkpoints, but using in together with bondage-v11 and merges of it might make things a bit easier.

Full list of trained keywords: https://pastebin.com/8nJSPM8N

I have a discord related to my projects, where I share things early, and help people with prompts and workflows.
https://discord.gg/MzEaugjyFA

If anyone wants to support me I have a Patreon and buymeacoffee page.

https://www.patreon.com/NotStrooge

https://www.buymeacoffee.com/NotStrooge

👑 Jr Carrington Lycoris👑

🔧Model base: AngrA_RealFlex_v2

__________________

📝🌟 Trigger: (JrCarringtonQuiron woman), <lyco:quiron_JrCarrington_v1Lycoris:1.0>, long hair, blonde hair,

__________________

📝 Variation: <lyco:quiron_JrCarrington_v1Lycoris:Y>, Y=0.8 to 1.0

📝 Denoising strength: 0.30 to 0.6

__________________

🗒️💡Sample:

(masterpiece, best quality, high resolution), (realistic:1.2), (JrCarringtonQuiron woman), long hair, blonde hair, <lyco:quiron_JrCarrington_v1Lycoris:1.0>

__________________

🚀 This LORA likes: (best quality:1.1), (masterpiece:1.2), (realistic:1.2), (detailed:1.1),

👇 Negative Prompt: (worst quality:1.2), (low quality:1.1), (normal quality:1.1), lowres,

💡Sample steps: 20 to 39

💡CFG Scale: 5.8 to 10.5

Hire. fix: ok

👇Restore faces: Yes /No

____________

☕ If you feel like supporting my work, please consider visiting my Beer / BuyMeaCoffe ( https://www.buymeacoffee.com/quiron ) ☕

🌟☕If you like my work please support ( https://www.patreon.com/VisionsofAi )☕🌟

A model featuring the spectral colors of light.

I'm still new to creating models, but I love it. If you like what I'm creating feel free to buy me a coffee. Good coffee is the best:

https://www.buymeacoffee.com/donmischo

>>Donate Coffee for Gtonero<<

if uses Conv2D requires need extensions
https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris
or
https://github.com/kohya-ss/sd-webui-additional-networks