自作の生成画像100枚で学習されたスタイルLoRAです。

This is a style LoRA trained with 100 home-generated images.

同時公開されているKataragi-testV1.0と組み合わせて画風を変更することができます。

It can be combined with Kataragi-testV1.0, which is released at the same time, to change the painting style.

光の質感が向上し、塗りが元のモデルよりブラシ塗りと水彩を混ぜたような特殊なものに変化します。

The light texture is improved, and the paint is changed from the original model to a mixture of brush paint and watercolor, and a special kind of brush paint.

图例用了 deepboys25D_v20 这个模型

练的第一稿LORA用了些网络上的骑士图片训练

建议参数<lora:MUknight:0.6>, armor, solo, 1boy, male_focus,

还有触发词wing和weapon

效果不是很理想逐步搞后面版本

城娘：Hakurojo-cg

私服：Hakurojo-pc

Google Translate:

castle girl: Hakurojo-cg

plain clothes: Hakurojo-pc

From Zentai Lover.

For Zentai Lovers!

Please note this LoRA is still a WIP project.

If you have any suggestion or bug report please feel free to contact me.

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

Base Model: ChilloutMix.

Zentai Girl LoRA model trained from Photos.

This LoRA will generate female characters wearing Zentai(Morphsuit) bodysuit. Could be combined with other LoRAs and CKPT models. All the ControlNet functions are supported.

Currently this LoRA is not tagged with Mature Contents since no Nudity exists.

recommended strength is 0.7-0.9

LoRA of Bradamante in FGO, works between 0.7-1.0

FGO屁股枪的模型，推荐权重0.7-1.0

-

Supports 5 sets of Official Outfits in total, 3 sets for Lancer, 2 sets of Craft Essences.

支持共计5套官方外观，L阶3套+礼装外观2套

-

Lancer1=Lancer1, 1st Stage Clothing

Lancer2=Lancer2, Closed Long Jacket

Lancer3=Lancer3, Bradamante‘s Bikini

Wellcome Bunny兔女郎=WellcomeBunny, Bradamante Bunny Girl

Heroic Spirit Festive Wear: Bradamante运动服= HeroicSpiritFestive, FGOFes Outfit

-

Feel free to post any idea or character model request in the Discussion

有模型的需求可以B站私信或者在这里留言给我

id：不吃青蛙的阿龙

有空就会做

If you like my work, feel free to support me on patreon.com/RiuKiMK1

or RiuKi_MK1 is Creating AI artworks, LoRa models (buymeacoffee.com)

ポケモンのカガリです　ちょい不安定気味　エンゲイジしたい人向け

Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord

Training settings:

35 images, 25 steps, 15 epochs, you find many epochs (with test images using plot xy script) on my Patreon while supporting me.

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)

• CFG Scale : 5 to 9

• LORA weight for txt2img: reccomended 0.2-0.7

• Keys: wearing flowers dress, wearing colorful dress

• Adviced Image Size: 512x768

This model is kinda hard to use, it's very important to start your prompt with something like "1 woman wearing flowers dress", don't place the keyword at the end of the prompt or you might not get what you want.

Also, i've tryied without my model and this keyword, many times it works even without the model, but the result is not the same. to give you a few examples, i made some xyz plot to show the same prompt with different weights on my model, you can find them in the images list.

You just need to practice and experience this model on your own, if you find a good prompt or you have any good advice i invite you to post and comment to give other people the chance to use it in the right way.

If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians

If you like my LoRAs please consider to support me by buying me a coffee here

And if you want to see my work diary and notice post of next LoRAs, click here

Model : chilloutmix_NiPrunedFp32Fix

Vae : vae-ft-mse-840000-ema-pruned

Please do not post NSFW or sexual pictures.

lora version of rimururu (SAMURAI SPIRITS) (Samurai Shodown)

sion(eden)

Common words:

[ aqua hair, brabulladbgt, red hairband, red gloves, blue eyes, skirt, hoop earrings, choker ]

Usefull Words

[ thighhighs, fingerless gloves, zettai ryouiki, thigh boots]

Trigger it with "Archi ZHS",

Lora created on the basis of screenshots from the anime "Bastard. Ankoku no Hakaishin". It turns out quite an interesting style from the 90s with a muted color palette.

Trigger world “ArchiZHS”

and also try another magic word "Lighton", put it in positive will trigger some light, and in the negative you will get a crved architecute without lights (likely )

A Reze Lora, not restricting the clothing style

model for sena bonbon of flavr, not good but first one ive made

If you like my LoRAs please consider to support me by buying me a coffee here

And if you want to see my work diary and notice post of next LoRAs, click here

Model : chilloutmix_NiPrunedFp32Fix

Vae : vae-ft-mse-840000-ema-pruned

Please do not post NSFW or sexual pictures.

这是一个制造3D 偏Q版可爱风格的人物Lora模型

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

IF YOU YOUR PICTURES HAVE PROBLEMS CHECK MY FAQ SECTION: FAQs

RIGHT NOW IMAGES METADATA ARE MISSING AS I'M USING COMFYUI TO SIMPLIFY THE IMAGE GENERATION. I'M TRYING TO FIND A WAY TO FIX THIS, IF SOMEONE KNOWS HOW TO DO IT, I'D LIKE TO KNOW.

What's this LoRA

This is a LoRA trained on illustrations and painting inspired by prints on wood. My models are not trained to achieve the exact style of their source, but to change the general vibe of the image. They are also fine-tuned on anime models, so they may not work perfectly with realistic ones.

The images in the examples are created using only this LoRA and different models (mainly my custom):

• https://civitai.com/models/24149

• https://civitai.com/models/26332

• https://civitai.com/models/28322

• https://civitai.com/models/32022

You don't need to specify a word to trigger this LoRA.

The LoRA works also on NSFW (check out the examples).

Guides

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

• Beginner's Guide

• Prompting

• My Workflow

• Inpainting

If you want to create images with a high LoRA weight, I highly suggest you to check the tutorial.

Support

Do you want to get access to my LoRAs before everyone, and contribute to my next models? Check out my supporter tiers on Deviant Art:

If you want to support my work for free, you can also check out my music/art here:

• Bandcamp

• Instagram

• DeviantArt

Edit: 19.04.2023: Last Chance to DL it

This LoRA will be deleted today night - because bad reviews - Nobody can read ... sad so i stop work on it.

Its a Beta and people think is a paid AAA-LoRA ... so pleace train yourself.

Version Beta 2 Changelog:

• Rework of Images / Caption

• Use Low Strenght ( Preview at 0.5)

• With other LoRA Compatible! (Use Low Strenght and search perfect strenght)

• All Preview are Simple Prompts!

• Adding flat_tits

6 different Breast Types (triggerwords)

• flat_tits

• small_tits

• medium_tits

• big_tits

• fake_tits

• bimbo_tits

Beta 2 Recommended strenght; 0.5-0.6!

Beta 1 Recommended strength: 0.6-0.8!

Work in Progress coming in next Updates:

1. bimbo_tits need rework

2. skin textures will be more realistic

3. Nipple Color + difference nipples

4. ethnic support

5. and more

small boobs note: small boobs pictures were trained by shemales and femboys as well as 21+ year old models

这是一个New Game! 主角凉风青叶的模型。

This is a model of the main character Suzukaze Aoba from anime New Game!

prompt里需要加上描述外观的提示词：purple eyes, long hair, twintails, purple hair, hair ornament, blush,

Just a model for training myself, don't expect too much because the art style of this artist is unstable.

Recommended weight:0.7~1.0

推荐权重：0.7~1.0

It is prohibited to use this Lora model for any commercial purpose.

禁止将此Lora模型用于任何商业目的。

It is forbidden to generate any NSFW pictures with this lora model.

原则上禁止用此lora模型生成任何NSFW图片。

¿Looking for Tzuyu nsfw version?

**Important read**

This model is a early version, Gradually I will add more pictures and updating the model until make it better and more complete

Hello here some settings that can help to get some good results

Recommended resolution: 512x768

• lora weight: 1.0 - 1.1

• Simpling method: DPM++ SDE Karras or Euler a

• Steps: 30

• Model: "Chilloutmix_NiPrunedFp32Fix" or "Chilloutmix"

• CFG: 7

You can use some trigger words like: tzuyu, chewy

also you can use: "blonde hair", "brown hair" to get some variations

不知道

这个模型会一直更新下去，会把优衣的几套衣服全都做完

示例图里的tag是几乎一样的，用的不同的大模型

好想换张显卡啊

城娘：Edoshiyakata-cg

私服：Edoshiyakata-pc

Google Translate:

castle girl: Edoshiyakata-cg

plain clothes: Edoshiyakata-pc

No.1xx 是一些我完全不认识的人，个别知道的我或许也因为懒得改名就放在这个系列里。

No.2xx 大概会是我个人喜欢的明星或者博主。

lora强度推荐0.6-0.8。chilloutmix出图，词条看示意图，负面提示词我用了EasyNegative.pt：EasyNegative | Stable Diffusion TextualInversion | Civitai

素材来自网络视频截图

请及时下载保存，真人模型随时可能被删除。

喜欢请给五星。

公共频道：t.me/dawn66666666

It generates Tiki aesthetics.

Perpetual coffee request: I signed up for Buy-Me-a-Coffee and Ko-fi, mainly because I thought it would be hilarious if someone as artistically untalented as myself received a coin for something creative. It's already happened, and it made my day. AI made the impossible possible :D. If you'd like to add to the amusement, please consider buying me more coffee through the following links:

https://www.buymeacoffee.com/konyconiN

https://ko-fi.com/konyconi7992

this lora try hidden nipples, but is failed lora, hope someone can help or make a better one

```

breasts, topless, (<lora:nonipples-000002:1:ALL0.5> nonipples)

breasts, topless, (<lora:nonipples-000002:0.5> nonipples)

```

```

EasyNegative (normal quality, low quality, worst quality:1.4), jpeg artifacts, (username,watermark,signature,time signature, timestamp, artist name, copyright name, copyright), (loli, child, infant, baby:1.3), (bad anatomy, extra hands, extra legs, extra digits, extra_fingers, wrong finger, inaccurate limb), (African American, African:1.6), (tits, nipple:1.6)

```

Cecilia from Fire Emblem The Binding Blade, trained on animefull

Works best between 0.7-0.9 weight

We will comply with the Hololive secondary creation guidelines.

纯属整个活，复刻名场面。

anything-v4.5效果不错，推荐权重0.8-1。

推荐结合closed eyes, teeth,clenched teeth,open mouth,可以大幅度提高场面效果。

关键词只使用slapping的话，可以解锁更多姿势，但不稳定性也提高。

This is just a recreation of a famous scene.

The anything-v4.5 model works well, and I recommend a weight of 0.8-1.

I recommend using "closed eyes, teeth, clenched teeth, open mouth" to greatly enhance the scene's effect.

If you only use "slapping" as a trigger word, it can unlock more poses, but it also increases instability.

可以出各种像素小人人儿！可以配合其他lora来玩！也可以出像素场景！权重0.5就行！

You can make all kinds of pixel charactor! You can play with other lora! You can also create pixel scenes! A weight of 0.5 will do!

いろんなピクセルキャラが作れます！ 他のローラと遊べる！ ピクセルシーンも作れます！ 0.5 の重みで十分です。

A request from a Kofi donator
PK from Girls' Frontline

Keywords:
PK
PKSleeveless

Recommend 0.8-1 strength

If you've enjoyed my models, consider leaving a tip.
https://ko-fi.com/leontalon

Trained on NAI.

Recommended weigth: 0.8.

Feel free to share your result on the comments.

Jacques de Molay Stage 2 From FGO.

Recommended Weight is 1.

Cover eyes

CFG Scale 10+ Increase success rates .

Not supported in real-world scenarios .

【Attention】 that if a single keyword does not work, enter it in the form of this sentence

EX: Strips of cloth, covering the eyes,Cover eyes with ribbon,1girl,

<lora:cover _eyes:0.6>,

挡住眼睛

提示词相关性提高到10左右，

暂时不支持现实场景，当然你也可以试试你的模型。

【注意】如果单一关键词不起作用，可以按我这个例子来输入：

EX: Strips of cloth, covering the eyes,Cover eyes with ribbon,1girl,

<lora:cover _eyes:0.6>,

【如果您请我一杯奶茶的话就更好了！】

【Thank you for your support！】

I'm bad at prompts

It's a little hit or miss on the results. Weight of 0.6-0.7 worked well with me. I was able to combine this with other loras with decent results.

Using this with anime models is possible, but far trickier for my smooth brain. I used the img2img DeepBooru Interrogator on a realistic image to come up with a prompt. Take a look at my prompts for some ideas. It's definitely more janky with anime Character Loras/ti.

Jasmine movie Aladin disney work with 0.6 weight enjoy it

triggers words: jasmine

follow me on Instagram @creativehotia

Trained in Anything V4.5 but should work fine in any anime model.

Lora weight: 0.6 ~ 0.80

Upscaler/settings I use

Highres: 4x animeSharp

Hires steps: 20
Denoising strength .55 ~ .75

Doraemon | ドラえもん

It is really hard to prompt

I hope some people can use this lora to make a good doraemon

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

This was requested through Google Forms

Weight: 0.7 - 0.8

Trigger + Appearance: hakodate omiko, scope

Default Outfit: visor cap, dress

Depending on your model, her scope can either show up, look weird, or not show at all. She can hold a (regular) tennis racket if you're lucky.

I created a Lora based on a dataset of 72 images of black and white hearts in Photoshop, just to see what happens. Got some interesting results.

Trigger word is: BWHeart.

Also here is the list of tags that appeared in the captions, so using these words in the prompt or negative prompt could also have a substantial effect on the outcome.

Models react to the Lora quite differently, and some require less power to make anime characters, while some require more to bring out the Lora's abstractness. So hard to give instructions.

It also has a tendency to create multiple girls, even with "multiple girls" and "multiple views" put into the negative prompt. But this makes it pretty good at making yuri images too!

If you like this one, you might also like my other Lora that is similar to this here: https://civitai.com/models/41046/bwgeoabstract

Created by GeomanticArts

Harle definitely hasn't gotten enough love. Lets fix that.

Virtually all art of Harle is in the same outfit, unsurprisingly it takes pretty high weights on clothing terms to change the clothes fully.

Example images all made with AOM2-7thV3_C_40 at weight 1

'Harle (Chrono Cross), facial paint, red leotard, see through blue harem pants, jester cap' example prompt to get you started

LoRA metadata has all the tags used.

My (nsfw) Twitter

Trained on NAI.

Recommended CHECKPOINT:revAnimated_v121.safetensors

If it is a female character you can try:ChilloutMix

You can also try other checkpoints, which may have unexpected effects

3D LoRA :OC

3D LoRA has no male characters

Lora trained on 188 images by Rilex Lenov.

I release all my models for free, if you feel like supporting me and my work you can support me here and buy me a drink.

Trained on NAI, gen'd on AOM2 hard.

The Princess of Dreams, Angel of Light, Student Council President and Heroine of Justice, Hinagiku Katsura!

The girl who was in top 10 best female characters in late 2000s, couldn't be none other than my favorite female character of all the time.

I exaggerated the amount of images for training, while it needed very low epochs to be decent, apparently Weight 1 is too much tied with the official anime's style, so you may need to reduce it for different styles, 0.5 or 0.6 works well.

Triggers with hinagiku katsura, and hakuou uniform brings her classic uniform. Supposedly summer uniform brings her white uniform but doesn't work very well, perhaps due to lack of many examples.

Trained on 88 fanart images

If i ever get to all the Alternian troll girls I'll do a single lora trained on all of them, as well as one for the boys, and then one that contains all of them (to increase compatibility of using multiple characters, since using multiple loras can be finnicky)

Mini Cooper Countryman trained on 768x768 images

Model strength 0.8 is ideal, but do experiment. For this one, precision of body color is not always accurate

A request from a Kofi donator
SL8 from Girls' Frontline

Keyword: SL8

Recommend 0.8-1 strength

If you've enjoyed my models, consider leaving a tip.
https://ko-fi.com/leontalon

High-quality models, come and download and use them！

优质模型，快来下载使用吧！

QQ交流群：816165884

JK uniform，It is prohibited to be used for any public figure model training

JK制服，禁止用于任何公众人物模型训练

The ideal weight is 0.4~0.6

推荐权重0.4~0.6

Testing only

This is not a necessary, but If you like my Lora, please consider to donate or even Comissioning your own requested Lora on My Ko-fi

https://ko-fi.com/warrensky

https://ko-fi.com/warrensky/commissions

Please give 5 star reviews by sharing your picture creation if you like it,

• Use any checkpoint based on SD1.5

• its 768,768 image

• turn on Hires.fix (or not, please experiment with it yourself)

• Hires steps: 14-17, denoising:04-65 ish.. just try it yourself

• CGE: 5-7

• Use DPM++ 2M Karras or DPM++ SDE Karras

• Use 20-ish sampling steps

• Strength: 0.6-1

• I suggest turn off the restore faces check mark for better results of this Lora

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

Legal & Risk

It is prohibited to use this model for commercial purposes and any scenarios of illegal acts and purposes.

本人原则上不支持使用此模型生成任何成人内容。用户在使用本模型时，请务必遵守当地法律法规，勿侵犯他人的名誉权、隐私权或肖像权。

I generally do not support the use of this model to generate any adult content. When using this model, users should abide by local laws and regulations and shall not violate any other's right to reputation, privacy or portrait.

私は一般的に、このモデルを使用して成人コンテンツを生成することをサポートしません。 このモデルを使用する場合は、ユーザーは当地の法律や規制を守り、他人の名誉権、プライバシー権または肖像権を侵害しないようにしなければなりません。

color plan

>>>GUIDE<<<

This LoRA is only for inpainting eyes.

First attempt at training a Lora. I had very few images to work with, but overall it does fairly well at capturing the character. In some models and prompts, it will lose either the hair or the eyes and you'll have to add them back. I tend to use "aqua_eyes" and "purple_hair."

I've had decent results around 0.7.

NSFW tag because nudes were used in the training.

Lora for amy from doaxvv. This was trained with 饭特稀_v0.8 which you can find on civit. The lora should work ok on similar models but it looks best on 饭特稀_v0.8. Trigger word is "amy".

Not trained on any particular costume.

NOTE: The training set had handful where one of her pigtails was covered by her head, so the model seems to get confused with her hairstyle, and likes to give her a ponytail. I see it just giving her long hair a lot as well. I might go back and retrain the model on better images at some point. But for now I think it gets her face and swept bangs pretty well.

Also highly recommend EasyNegative which my sample images are using.

Brunnya from Fire Emblem The Binding Blade, trained on animefull

Works best between 0.6-0.9 weight

0.8-1.0 weight

trained on anime screenshot

使用权重为0.8-1.0

基于动漫截图训练

20 pictures of soldiers 76 of 512*640 were trained.

!!!!!WARNING!!!!!! This Lora is in Alpha mode. It may not work properly. But I need testing to improve it !!!!!!!!!!!!!!!!!!!!!

Are you afraid of ghosts ? Yes ? Well.....................

Do you like groping ? No ? Well...................................

Do you think cursed armors/items are the plague of RPG's ? No ? Well............................

Well, you came accross this Lora by miracle then. It allows you to summon pervy ghost disembodied hands to grope cute girls on their most private parts. Good thing to make some cursed items.

Recommended weight : 1, if below, less hands will appear. There may even be no hands at all.

Recommended prompts : grabbyghosthands, gropped by multiple hands, (breasts groping), (pussy fingering), disembodied hands

Resources for negative prompt :

https://civitai.com/models/30452/bad-hands

https://civitai.com/models/5224/bad-artist-negative-embedding

https://civitai.com/models/4629/deep-negative-v1x

https://civitai.com/models/11772/verybadimagenegative

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

Trained with Defmix : https://huggingface.co/Defpoint/Defmix-v2.0

I was able to use it with other models, but it's very gacha.

I am the Reindeer Czar, so I created my second LORA, of Red-Nosed Raider from Fortnite. This depicts her really well I think.

Version 2 improves the model very much. No more issue with background, overall more expressive and high quality. 0.5 - 0.7 weight for LORA is good. Euler a and DPM++ seem to be best sampling. Sample around 50 seems good too, more can be better too.

You can add tags like sitting, open mouth, smiling, holding weapon, lightsaber, smirk, candycane, shotgun to use for some dynamic situations.

If you like these, please consider a tip, to help me train more models!

https://ko-fi.com/reindeerczar

Deprecated V1 Description:

Strength for LORA is say, 0.4 to 0.65 to get the good result, if you want a background.

It was trained on 5850 iterations, so quite much. Add the tags blue gradient background and simple background in the negative tag to get rid of the background better.

I'm open to Character + Style requests!

Description is in a work in progress

I'll begin working with the thumbnail if I have sufficient enough characters.

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

Hyoma Chigiri (4/18/2023)

Weight: 0.8

Trigger + Appearance: chigiri hyoma, long red hair, red eyes

BLUELOCK Uniform: sportswear

Seishirou Nagi (4/17/2023)

Weight: 0.7 - 0.8

Trigger + Appearance (Optional): nagi seishirou, white hair

BLUELOCK Uniform: sportswear

Yoichi Isagi (4/17/2023)

Weight: 0.7 - 0.8

Trigger + Apperance: isagi yoichi, black hair

Outfits: He's either wearing a bodysuit or blue sportswear.

Meguru Bachira (4/13/2023)

Weight: 0.8

Trigger: bachira meguru

His short, multicolored hair and yellow eyes are pruned, but feel free to add them anyway.

BLUELOCK Uniform: The uniform will very likely not appear when generating a wider shot, but throw him some sportswear, I'm sure he'll wear it.

The focus is on grabbing the ass
It is recommended to use anything that fixes the big problem with the hands and eyes...

grab,<lora:graa-15:1> --- Activacion

ass (76)

ass grab (75)

hetero (73)

1girl (69)

blush (58)

groping (57)

solo focus (55)

1boy (50)

breasts (49)

skirt (45)

underwear (44)

train interior (43)

panties (41)

chikan (40)

long hair (38)

short hair (37)

clothes lift (36)

school uniform (34)

molestation (34)

looking back (33)

skirt lift (32)

brown hair (29)

multiple boys (28)

large breasts (27)

from behind (26)

grabbing another's ass (26)

open mouth (24)

shirt (23)

public indecency (22)

2boys (22)

white panties (21)

black hair (20)

thighs (20)

sweat (18)

ass focus (17)

huge ass (17)

pussy juice (16)

penis (16)

pleated skirt (16)

smile (15)

blonde hair (15)

closed eyes (15)

censored (15)

fingering (14)

deep skin (14)

ponytail (14)

bangs (13)

hair ornament (13)

dark skin (13)

panty pull (13)

blue eyes (13)

pussy (13)

serafuku (13)

thighhighs (12)

jacket (12)

pantyhose (12)

thick thighs (12)

Tzuyu (TWICE)

I'm working on my LoRA technique so here is Tzuyu in all her glory. Please let me know how this looks!

Do you like my work? Buy me a beer!

If LoRA isn't your thing you can still download my Textual Inversion of Tzuyu HERE

Velouria from Fire Emblem Fates

This was my first LoRA, and i didnt knew half of what i know now, so i decided to remake it

Use velouria_fates as trigger word.

Trained on Nai

You may want to add "red eyes, gray hair, multicolored hair, animal ears, red hood" to your Prompt

Can be used at 1.0 weight but you should lower to 0.7 or 0.8 it if you want to change her clothes and maybe you will need to add "belt, buckle, corset" to your negative prompt as well

All images were generated using yabalMix (https://civitai.com/models/28648/yabalmix) with xformers on and using kl-f8-anime2 vae.

images were generated at 512x768 and upscaled using highres fix by 1.95 and using 4x_foolhardy_Remacri upscaler

Janice Griffith

She left us too soon...indeed she's left the industry and is now living out her life as one of the most beautiful goddesses to grace our eyes.

I'm practicing my LoRA creation, I hope you like it!

Do you like my work? Buy me a beer!

If LoRA isn't your thing, you can still download my Textual Inversion of Janice HERE

Zendaya

Actress and model, also has quite a future in show biz. She was on a Disney show, I believe, so I hope that doesn't come back to bite her.

I'm practicing my LoRA creation, I hope you like it!

Do you like my work? Buy me a beer!

You can still download my Textual Inversion of Zendaya HERE

Lora of nijisanji en Petra Gruin

Lora of nijisanji en Selen Tatsuki

Lora of nijisanji en Reimu Endou

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

This was requested through Google Forms

Weight: 0.6 - 0.8

Trigger + Appearance: kunieda aoi, blue hair, necklace

Default Outfit: open clothes, white coat, (sarashi:1.1), white pants, midriff, cleavage

Casual Outfit: kunieda aoi \(casual\), hat, glasses, semi-rimless eyewear, under-rim eyewear, pink jacket, yellow skirt

My training dataset consists mostly of screencaps, so increasing the weight will give you lower but more nostalgic quality. I highly recommend hires fix. You may swap blue hair for black hair.

Lora of nijisanji en Enna Alouette

Hare, from Blue Archive. Use at ~.8 strength

Wiggle Wiggle Wiggle 🍑

The examples shown from this LoRA have post processing (hirez and inpaint).

See the original prompts ver. on gallery.

See also: Veiny Breasts LoRA, that you can found in my Patreon. (only for Longer Prompters+)

LoRA prompt weight: 0.6 to 0.8
Model used for training: Uber Realistic Porn Merge 1.3.

Recomendations:

• CFG ~9

• Steps >=27

• DPM Family or DDIM

Prompts that can improve your results: cellulite, stretch_marks, thick_thighs, huge_ass, round_hips, square_hips, slim waist...

Tendencies of this LoRA (and resolution):

• The LoRA changes faces a bit, you can inpaint, hirez-ing, or put some prompts to improve the face details

• This model also fattens girls, just put plump, fat and curvy on neg and slim on pos.

• This model can work with underwear or not. To nudify use: (underwear:1.4), (panties:1.4), (swimsuit:1.4), (leotard:1.4) on neg. and nude in pos.

See my other LoRAs:

• Puffy Nipples

• Perfect Perky Breasts

• Tight Bikini Top

• Big Nipple Tip [Lite]

• Small Breasts [Lite]

• Big Natural Breasts Type01 [Lite]

• Big Natural Breasts Type02 [Lite]

• Ghost Nipples [Lite]

• Underwater Naked Shot

• Fake Breasts [Lite]

• Macromastia [Lite]

🐔

Mary Kish is a twitch streamer and personality in gaming.

expression chart for a chart of expressions

expression chart, full body for a chart of expressions and a full body image

works particularly well with the high res fix, if you don't upscale, then a lot of the time the expression parts will be too low res.

Combines well with character loras.

Trained on a bunch of different types of expression charts, what you get is mostly gacha; the only control you have is in how many different expressions you include as tags, and if you use full body or not.

Demo images use Danger Zoo ~ Johnsons Negative TI Collection ~

"modelname style" is the main trigger word
additional trigger words are listed to keep you close to the trained style/concepts.

This LORA is designed for used with other djz LORA's and/or on top of Mixed pairs of djz CKPT's or a Triple merge mix including other CKPT models. This is a specific method for merged CKPT models while running a stack of these Style LORA models on top. You may still use this alone and it is recommended to include the additional trigger words in your prompts.

DJZ Season 2.5

Li yitong, is a young and beautiful Chinese actress. She is lovely. Do not create or share any NSFW image from my LoRA.

Atelier Totori - Totooria Helmold トトゥーリア・ヘルモルト LoRA

Trained mostly based on official artworks.

Recommended CFG: 5-6

Recommended weight: 0.8-1 (depends on your personal settings)

Recommended prompt: blue sleeves, detached sleeves, skirt, dress, bare shoulders, blush, collarbone, strapless, hat, frills, flat chest, bow, back bow,

NOTE: if you do not want to get ribbon or bad back bow generation results, removing bow, back bow from your prompt can help you.

I used to train 7 version of Totori LoRA, this LoRA version is the best one.

The reason her LoRA is so difficult to produce is largely due to her clothes. With a limited training sample, you'll have a hard time finding quality clothing details to use as a reference. So I had to spend more time on tag editing. I tried generating it several times and it has a relatively good performance at 768*1024 resolution.

Trained on 102 images off of her Instagram.

It is Lona from Lona Rpg, if you know you know

I would recommend using at 0.6 strength

The trigger word is lona

Do not use dark skin, dark-skinned female

Trained on images of human Lona with all clothes, chains and piercings

W.I.P. I still need 1. images of different race Lona 2. pregnant Lona 3. Damaged Lona 4. sex acts i.e. vaginal, anal, dp, pissing, defecation

This model was created with 100 images of the animation directly, you can create She-ra and Adora, but the sword only with controlnet.

I recommend to use it with 0.5 to 0.7 of strength

To generate Adora use these tags:
1girl, blonde hair, Adora, rating:safe, solo, red leotard, leotard, belt, long hair, breasts, medium breasts, bracer, wristband, makeup, lipstick, 1980s \(style\), retro artstyle,

To generate She-ra use these tags:

1girl, solo, She-ra, long hair, blonde hair, blue eyes, tiara, white dress, choker, red cape, makeup, lipstick, miniskirt, boots,

breasts, medium breasts, bracer, wristband,

rating:safe, 1980s \(style\), retro artstyle,


For the Pegasus Swift Wind:
weapon, sword, cape, parody, Sword of Grayskull, sword of power,

white horse, riding, pegasus, unicorn,

This is my first publicly released LORA! Simone was trained on 25 generated AI images using a celebrity mashup prompt. I would recommend using a weight of 0.7 to get the best results!

Simone should have blonde hair and blue eyes naturally, but she can easily get a makeover with good prompting!

If you're experiencing issues with generating images that are coming out deformed, feel free to see my guide here on fixing that issue and upscaling your images: https://ko-fi.com/post/The-Double-Headed-Issue-E1E0JO2UV

If you're unable to create the exact image, please read this comment section here: https://civitai.com/models/14065/faetastic?commentId=69125&modal=commentThread

All images were generated using hi-res fix. Feel free to ask me any questions if you're having issues, and I am also available on the CivitAI discord. :)

Made for a request by another user.

v1.1

Use strength 0.7-1.0, work in progress, has a problem with hair braid not vary accurate.

v1.2

Use strength 0.6-1.0, use braided ponytail, for better hair results

V1.3 use strength 0.55-0.8

V1.4 made. same as v1.3 but with leotard images

Trained with some NSFW images. first attempt was with a lot of images this is the 3rd version with only a few images. trying to figure out how to balance them, like my other lora no Triger words just try to describe her

A lora based on the female body type venus body

Its a body type that has thick thighs,wide hips,(and usual giant breasts). combined with having a somewhat slim waist. Without the girl being fat and having a massive stomach.

I found out that it was quite hard to get this bodytype in most models especially realistic so I made a lora for it

Trained with 301 images

tested on 10 models

Image info

• Model used kotomix

• Model used abyssorange2hard

• bad hands embedding

• bad artist embedding

• bad prompt embedding

• deep negative embedding

Notes

• I didn't prune any tags so you have to specify what you want ig(gigantic breasts,thick thighs,gigantic ass) etc

• You can get pretty extreme proportions if you widen the resolution (640x640)

Known Issues

• using this lora on realistic models has a higher chance for a double belly buttonbelly abnormalties at higher weights

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

1.5, seems to work best at 0.7-8, might have to pump up to .9 on certain checkpoints. Seemed to pair well with a few other loras i tried it with.

adding "with weapon" will add rocket launcher usually but will probably need some inpainting.

will try to fine tune and update as i progress in lora making. going to try Tyrant next. let me know any characters you'd like to see.

support me if you want : https://ko-fi.com/toadviner

My fourteenth Overwatch LORA. Original appearance.

Main trigger\Face - ashe, red eyes

Outfit - cowboy hat, necktie, suit, jacket, gloves, shoulder amor, high heels

Checked strength range is 0.6-0.8.

Making and testing LORA's takes quite a bit of my time, so I would appreciate if you bought me a coffee.

Inpaint guide: if face on full body shots comes out deformed\not detailed enough - it's easily fixed with single inpaint, whole process takes half a minute. Press "Send to inpaint" button under generated image, mask the face area with drawing tool, choose DDIM sampler, set "Inpaint area" to "only masked", and set denoising strength to 0,5. Press generate and you done!

Review and comment with your experiences, I would appreciate whatever feedback you can provide in order to continue improving my technique.

Trained on Anything V4.5 with anime.vae.pt and CLIP Skip 2.

Samples are 512x768, without HighRes.fix, using OrangeCocoa 50/50 Mix , with OrangeMix VAE, and using EasyNegative embedding, CLIP Skip 2.

Hide

Made this loRA, one of my first ones ever so its not great. It creates some artifacts on the neck area (don't know why).

All images trained on are close up AI upscale photos from her insta. This is also the reason it looks more 2.5d than photorealistic as the AI puts smooth filter over face when upscaling.

This is a LoRa model trained from 400+ images from anime Oshi no Ko.

You can use the LoRa to design roles.

Weight 0.9-1.0 will perform better

The first episode is very long...

We will comply with the Hololive secondary creation guidelines.

Ainz Ooal Gown アインズ・ウール・ゴウン

https://www.pixiv.net/users/86166786

地点：(Night:1.7), in dark castle,purple sky,Cherry blossom rain,magic_circle, medieval ,
动作：standing in front of a magic_circle,
服饰：shoulder_armor,armor, cape, cloak, grin, hood, hood_up, hooded_cloak, king_hassan_\(fate\), male_focus, mask, skeleton, skull, black_armor, cape, fake_horns, glowing_red_eyes,
角色：<lora:AinzOoalGown-V1:0.85>,<lora:cyberpunkStyle_v10:0.3>,<lora:AuroraTheEminenceInShadowLORA_v4:0.1>,<lora:jokerArkhamverseLora_v1:0.1>,
角色鎖定：
a skeleton with a helmet and a sword in his hand and a helmet on his head and a sword in his hand, 1boy, glowing, glowing_eyes, hood, looking_at_viewer, male_focus, red_eyes, skull, solo, teeth, upper_body
a skeleton in a skeleton suit holding a sword and a skull in a crown in a stadium with a stadium in the background, armor,
畫質:
official art,extremely detailed CG unity 8k wallpaper, perfect lighting,Colorful, Bright_Front_face_Lighting,
(masterpiece:1.0),(best_quality:1.0), ultra high res,4K,ultra-detailed,
photography, 8K, HDR, highres, absurdres:1.2, Kodak portra 400, film grain, blurry background, bokeh:1.2, lens flare, (vibrant_color:1.2)
Negative prompt: EasyNegative, bad_prompt, bad-artist-anime, bad-hands-5, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, Missing vagina, Blurry faces, Blank faces, bad face, Ugly, extra ear, amputee, missing hands, missing arms, missing legs, Extra fingers, 6 fingers, Extra feet, Missing nipples, ghost, futanari, Extra legs, Extra hands, panties,pants, (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), text, error, blurry, jpeg artifacts, cropped, normal quality, artist name, (worst quality, low quality:1.4),
Steps: 27, Sampler: DPM++ SDE Karras, CFG scale: 7, Seed: 1412203222, Size: 512x768, Model hash: 4199bcdd14, Model: revAnimated_v122, Denoising strength: 0.5, Hires upscale: 2, Hires steps: 18, Hires upscaler: R-ESRGAN 4x+

masterpiece, best quality, helltaker <lora:Helltaker:0.7>

Negative prompt: lowres, (worst quality, low quality:1.3), (3D:1.3), signature, watermark, username, artist name

Size: 512x768, Seed: 3365297045, Model: AbyssOrangeMix2_nsfw-pruned, Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 11, Clip skip: 2, Model hash: 264bff1fbd, Hires upscale: 2, Hires upscaler: R-ESRGAN 4x+ Anime6B, Denoising strength: 0.5

This is model uses the exact same input as bimbostyleThree but was trained on deliberate_V2 which hopefully means that it's better applicable to other .ckpt in comparison to the normal "Three"

Since i mainly use "draw style" ckpt i can't really say anything about the full potential of this Lora on other ckpt.

Also if you want to mix my models, i would recommend to always putting the stronger weighted Lora in the front. i.e.

<lora:bimbostyleThreeU:0.5>:<lora:bimbostyleTwo:0.3>

This is an additional and more refined version of the second one.

I added different models and changed the learn prompts a bit.

This Model is trained on revAnimated_v11 and works probably best there.

Also if you want to mix my models, i would recommend to always put the stronger weighted Lora in the front. i.e.

<lora:bimbostyleThree:0.5>:<lora:bimbostyleTwo:0.3>

C18 v1

All my work with all prompt here (include NSFW)

patreon.com/Beerus_geek

This model aims to reproduce the unique art style seen in the eyecatches from the 2007 anime "Tengen Toppa Gurren Lagann".

Important!

For the best results apply the LoRA on checkpoint AnyLora

Try strengths: between 0.5-0.9

If you're looking to reproduce some of my examples, don't forget to update the LoRA filename in the prompt. These were from training and I forgot about them.

For more details see https://github.com/jamesz96/ttgl-eyecatch-LoRA

Alisa Selezneva, trained on 50+ (with some black and white)

I`ve got best results in this model https://civitai.com/models/4978

and <lora:alisa_selezneva:0.8>

Works best at 0.8

SXZ D.U.C.K. - Digital Units Creation Kit

All the content I make is and will be completely free on CivitAI.
But if you want to support me with a coin, then you can do it here.

Sample images was generated with 🌕Luma and AOM2

Recommended embeddings:
1. easynegative
2. bad-hands-5
3. badhandv4
Strength 0.6-0.8 is optimal.

PLEASE MAKE SURE THAT YOUR LORA FILENAME MATCHES LORA NAME IN PROMPT.

So basically this lora is a model with 5 different but similar styles. You may find it useful for creation game asset like icons, sprites and character portraits.

HOW TO USE:

Every style should start with icon of or icon portrait of, but every style also should have different ending prompt. But you may freely combine elements of this styles by mixing prompt like round frame with glow color and background color.

Also i recommend to use square resolutions like 512, 640, and 768, but you always can do some experiments.

In the end your promt should look like this:
1) icon of YOUR OBJECT, YOUR PROMPT, COLOR glow, COLOR background
2) icon portrait of YOUR CHARACTER, YOUR PROMPT, COLOR glow, COLOR background
Let me show you.

Style 1: round frame, cyan (OR ANY OTHER) glow, brown background

Style 2: blue glow, purple background

Style 3: white background

Style 4: dark background, gradient background

Style 5: round frame, blue glow, purple background

Enjoy and share your feedback.

Classic Hindustan Motors Ambassador made in India.

Now ambassadors can also travel the world

0.6 - 0.9

I made so many versions of it that I don't remember how it was trained, it should work the same way as the futanari horse penis, because it has a wide range of furry images it tends to do it without being prompt, if it has a lot of animal parts try to put in the negative prompt.

The futanari_canine_penis, it can be over trained, I recommend using it in 0.7,lower epochs should work fine,

Warm welcome to the newest apparation in the Black Bullet roster : Kayo Senju

Use at 0.8~1.0 weights

The trigger word is "bbkayo"

Due to her really low screentime and scenes, most of the dataset is biased with shadows of the "bunker" scene. you will need to add the "blonde hair" tag as her hair will be a little "bleached" by the training.

Tested on :

• Greymix

• MidnightMelt

• AOM3

• Kotosmix

• Sweetmix

• Mandarine

If you have any questions feel free to ask.

Kaïros#4463

Enjoy ♥

Lora Model of Alexandra Daddario

used 455 pic dataset

Because the site refused to let me upload two LoRA files, HL2Beta002 is zipped and categorised as training data.

The dataset for this one is fifty images taken from the concept art drawn for Half-Life 2.

I created this LoRA because I'm a huge fan of 2000s art with colourful lighting and complex backgrounds, and I hate how the average AI image has bright white lighting with solid colours passing for walls in the background.

Features:

• Great skies - especially cloudy and non-blue skies

• Great lighting

• Great pipes and sci-fi machines

• Good "sense of space"

• Good at low-brightness scenes

• Occasionally uses blur for an authentic illustrative effect (mostly around distant windows)

Limitations:

• Not good at bright blue skies since the dataset basically had none - rely on your base model for quality there

• Can kind of be "crowded out" by a strong character prompt and have no effect - in that case, you need to add a strong background prompt as well

Usage Instructions:

This download has two files - HL2Beta001 and HL2Beta002.

They're mostly identical, but HL2Beta002 is purposely a little more deepfried than HL2Beta001, since I was disappointed with 001's results.

What I like to do is use both of them in combination - HL2Beta001 with a weight of 0.5 and HL2Beta002 with a weight of 0.5.

If you want a stronger result, use only HL2Beta002 with a weight of 1.

If you want to mix it with another LoRA, use only HL2Beta002 with a weight of 0.5.

Nature images like to place whatever you want to generate on top of really flat, horrible cliffs. To mitigate this, you might like to put ((quarry, pyramid)) in your negative prompt.

Common keywords:

• Half-Life 2 Beta - shared with all images, good "trigger keyword" for the LoRA

• HL2 concept art - same as above

• city 17 - shared by all images of City 17 - this means green cloudy skies and skyscrapers

• industrial - the more industrial parts of City 17 - use "industrial window" and "brick factory" for a factory-like area

• technology - shared by all images with pipes, Combine technology, cables, etc... I find it more effective to just write "complicated network of pipes"

• air exchange - shared by all images of the Air Exchange - red skies and sunlight with blue shadows, lots of pipes

• Citadel - tall metal tower, seems to be unreliable in practice

• Combine - shared by all images of Combine technology, seems unreliable

• palace - tries to generate tile floors and blue light, seems unreliable

• depot/wasteland - tries to generate railroad bridge and big metal building on top of a cliff

• quarrytown - shared by all images of beta Ravenholm, which at the time was just a sunny mountain town

• seafloor - tries to generate a rough, bumpy wasteland - too small sample size

• nighttime, midnight, dark clouds - gives you a darker sky

• grey-blue sky - gives you a faded blue sky with lots of tiny clouds

• grey-green sky - gives you an apocalyptic green sky with lots of tiny clouds and sometimes a splash of yellow

• grey-red sky - gives you an apocalyptic red sky with lots of tiny clouds

• grey-yellow sky - can give you a nice sunset or a horrible piss-yellow sky at random - use "clouds lit from above by evening sunlight" or "clouds lit from above by early morning sunlight" in your prompts instead

• grey-orange sky - same as "grey-yellow sky"

About the preview images:

The preview images are kind of disingenuous. They're img2img outputs, and basically 1024x1024 recreations of images I generated at 512x512 with txt2img and a different seed. So, it'll be very hard to get an exact recreation. I'll try to leave a comment down below with the 512x512 originals, so you can at least get the seed for both stages. Any images that don't have an equivalent image in the comment already are 512x512.

I'll also try to leave a few comments with different models, since all of the preview images were generated with dreamshaper_4BakedVAE.safetensors.

From my limited testing so far, it seems like other models have similar results, but with different building shapes.

Note #1: This was probably the hardest character Lora that I've done so far as getting those stars to appear ended up making me bake the Lora more than I would like in general. And then, surprisingly, her bodysuit was even worst. For some reason it would randomly not show the whole thing and giving a midriff in more pictures than I'd like. This ended up causing me to make it more rigid than I generally like (though you can still change outfits and more as shown in my preview pictures), you will just have to compensate with a lower weight and adding a couple tags likely. I'll need to do a v2 eventually for this.

What does it do you ask? I just started playing through Fire Emblem Engage and hadn't truly liked a character yet in the game... Until I met Yunaka. Love her visuals as well as the fact that she actually has a personality that isn't super one-note or bland. So, of course, I had to make a Lora after playing with her lol.

How to use it:

Since this Lora is more rigid than I like, weights for changing around her outfit will likely be 0.6 to 0.7. And sadly you will lose some of the stars and the star on her face when you do that.

(I use Sampling Method DPM++ SDE Karras and Upscaler R-ESRGAN 4x+ Anime6B)

Main Triggers

YunakaFE - Main trigger word.

Bodysuit - If you are having trouble with the bodysuit being fully connected.

Recommended when changing outfits

red hair, star hair ornament - These help keep her visual quirks.

Optional

Cape - If you want to make sure the cape appears.

Extra, Extra!: Should be well trained and usable in most models. Images shown for example. More rigid than I'd like though.

Picture Descriptions:

1-2: AnythingV3 using just this Lora.

3: AnythingV3 combo'd with my in progress Demonic Transformation Locon.

4: AnythingV3 combo'd with my in progress Knight-ify Locon.

5: AnythingV3 combo'd with my in progress style of Defaultz17 Locon.

6: AnythingV3 combo'd with my in progress Hypno Harem Locon.

7: AnythingV3 combo'd with my released Pet Play Locon.

8: AbyssOrange using just this Lora.

9: PerfectWorld using just this Lora.

10: RevAnimated using just this Lora.

How was it trained?

It was trained on animefull_final with 26 images with the usual self-pruning and tagging and a wide range of different images. I like keeping my prompts simple for the first few pictures so that you can see it without a whole bunch of mumbo jumbo. Make your pictures better than mine though!

What's next for this?

Maybe I'll try Loha's out to see if they can solve the flexibility problem. Either way I need a v2 to tackle that at some point.

What's next that isn't this?

More character Loras. (Viper and Rupee from Nikke, Spectre Tenebria from E7, Zephia from Fire Emblem Engage, Ivy from Soul Calibur 5, Heine from Combatants Will Be Dispatched, Conqueror Lilias from E7, Evelynn from LoL (both base and All Out version), Tharja from Fire Emblem, Laharl-Chan from Disgaea 5, Miku from Idaten Deities, Kuribayashi Shino from GATE, Rin Sohma from Fruits Basket, Kasumi from P5, Ann from P5, Ikumi Mito from Food Wars, Alice Nakiri from Food Wars, Chika Fujiwara from Kaguya-sama, Lulu from FF10, Momo from My Hero, Sheik from Zelda, Lucoa from Dragon Maid, Laharl-chan from Disgaea, Vivy from Vivy: Fluorite Eyes' Song, Dominique de Sade from Vanitas, and Dos from One Punch Man.) And maybe Wilhelm van Astrea, my favorite Re:Zero husbando or a Prinny from Disgaea.

V2's of some of the previous Loras. Specifically make them more diversely acceptable with other models and LORA's.

I also have some more quite NSFW Locon's. I decided to list them as well, so if people have preferences on which to fix up first let me know. Specifically, Beta Censorship (Automatically censors any images you create to make them BetaSafe), PhonePic (Thisone has been very, very hard... But TLDR, someone holding a nuder version of the waifu on the phone), Demonic Transformation, being stepped on, Robotification, Defeated (decently NSFW defeat scenario for the waifu getting abused), Petrified (I'm unsure if I should bother with this one anymore), Infection (this one is VERY nsfw, as the waifu is completely infected), Dominatrix (waifu becoming a dominatrix), Brainwashing Machine, Tentacles, Pillory, Masked Evil (A mask that possess' the waifu), Latex Corruption, Combat Drone (hard to describe but basically making them a mindless sci-fi combat warrior), JAV Supervillain, Jiangshi (Transforms into a Jiangshi), Mind Flayer Defeat, Blacked, Hypnotized Harem, Hourglass Figure (just to make your waifu curvy), and Holding a Leash.

And, uh, I want to make a few SFW Locon's so I have an GoodCop/BadCop (Locon with outfits for being a good and for being a bad cop), Biker Chick (ride a motorcycle and look like you should), Race Queen (Chiling sexily on a car) Knight-ify, forest fairy, Anime Officer, barbarian.

Additionally, I love the style of Defaultz_17, so I'm making a Locon to mirror that.

Or just give me some random suggestion or priority opinions. That works too.

I've also added a couple of my extreme NSFW preview images to my twitter (https://twitter.com/TwoMoreTimes89). Let me know if you think I should work on them more for a submission here.

Nanase Ren 七瀬恋 | Night Shift Nurses

https://www.pixiv.net/users/86166786

地点：(Night:1.7), hospital, infirmary,Cyberpunk_CityView, Before Window,
动作：sitting at attention,sitting on couch,
服饰：(White_apron), dress,(pink bowtie),(puffy_short_sleeves),(pink shirt),(nurse_cap),(nurse_Outfit),
角色：Nurse,<lora:NanaseRenV1:0.7>,<lora:odaNonStyleLORA_v2:0.3>, oda Non,NanaseRen,
头发：green_eyes,long_hair,LOW_ponytail, white_hair_ribbon,aqua_eyes, (Closed_mouth),
角色鎖定：
1 girl,Solo, Manga,20yo,mature female,Beautiful Finger,Beautiful long legs,Beautiful body,Beautiful Nose,Beautiful character design, perfect eyes, perfect face,
眼神表情：
looking at viewer, (innocent_big_eyes:1.0),(expressionless:1.3),
畫質:
official art,extremely detailed CG unity 8k wallpaper, perfect lighting,Colorful, Bright_Front_face_Lighting,
(masterpiece:1.0),(best_quality:1.0), ultra high res,4K,ultra-detailed,
photography, 8K, HDR, highres, absurdres:1.2, Kodak portra 400, film grain, blurry background, bokeh:1.2, lens flare, (vibrant_color:1.2)
身材:
(Beautiful,large_Breasts:1.4), (beautiful_face:1.5),(narrow_waist),
Negative prompt: EasyNegative, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, Missing vagina, Blurry faces, Blank faces, bad face, Ugly, extra ear, amputee, missing hands, missing arms, missing legs, Extra fingers, 6 fingers, Extra feet, Missing nipples, ghost, futanari, Extra legs, Extra hands, panties,pants, (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), text, error, blurry, jpeg artifacts, cropped, normal quality, artist name, (worst quality, low quality:1.4),
Steps: 27, Sampler: DPM++ SDE Karras, CFG scale: 7, Seed: 249399667, Size: 512x768, Model hash: 3e9211917c, Model: yesmix_v16Original, Denoising strength: 0.5, Hires upscale: 2, Hires steps: 18, Hires upscaler: R-ESRGAN 4x+ Anime6B

Used embeddings: easynegative [119b]

Time taken: 1m 33.20s

Torch active/reserved: 10393/11260 MiB, Sys VRAM: 12288/12288 MiB (100.0%)

Lora trained on Ahri from League of Legends

models used

anime: Based65, AnyLoRA

2,5D: Lyriel, fantasticmix_real, ReV Animated

tags: ahri \(league of legends\), AhriCoven, 1girl, blonde hair, multicolored hair

optional tags: hair over one eye, cleavage, animal ears, solo, long hair, blonde hair, fox ears, bare shoulders, tail, fox tail, facial mark, magic, makeup, fingernails, looking at viewer, purple lips, bangs, sharp fingernails, parted lips, dress, lipstick, detached sleeves

negative embeddings: embedding:bad-hands-5, embedding:ng_deepnegative_v1_75t, embedding:bad-artist, embedding:EasyNegative

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• VAE: orangemix.vae (for 2,5D pictures), vae-ft-mse-840000-ema-pruned (for realistic models), kl-f8-anime2, blessed2.vae (for anime models)

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

• Steps: 20 to 35

• Clipskip: 1 or 2

• Upscaler : img2img in comfyui using 4x-UltraMix_Smooth

• CFG Scale : 5 to 9 (mostly 7 and 8)

• LORA weight for txt2img: 0.5-1 (for the examples i mostly used 0.7)

• Denoise strength for img2img: 0.4 to 0.7

If you want the exact same settings, just drag and drop the picture into ComfyUI

Trained with AnyLoRA checkpoint

for the pictures i used comfyui

IronCatLoRA #6 - Genos | One Punch Man

Making LoRA's is a thing I do for fun. So if you like what I do? You can help me out by just using the LoRA and leaving a review!

This time I present to you a character LoRA for Genos from One Punch Man. I like One Punch Man quite a bit. I especially think Genos is quite the badass. I noticed that there was also was no Genos LoRA to be found on CivitAI, so I just had to make it. I hope you're able to make badass images with Genos too, just like I was!

A weight around 0.8 for this LoRA seems optimal, usage may vary. From my testing a weight of 0.7 to 1.0 seemed to work good. Example images I've uploaded are mostly using 0.8. I've used YesMix for all the main example images, since in my own opinion it compliments Genos' look really nice (yeah kinda strange model for Genos, but I make big testing grids with multiple checkpoints, it was just the way like he was glowing looked really good). I've also uploaded some images generated with ReV Animated, which seems to generate semi-realistic images with Genos really well.

The main trigger word is genos. I tried to make this LoRA as flexible as I could, because of that you have certain additional words you must use to get certain characteristics of Genos. You can take a look at the example images I've uploaded how to prompt him, but here is an overview as well:

• mechanical arms, if you want to make sure he generates with his mechanical arms add this in your prompt. He will have them most of the times, but you can't go wrong by just adding it if you want his mechanical arms to be in the image.

  • You could also e.g. put it in the negative prompt so you can more easily prompt Genos with long sleeved clothing. Which may generate him with normal flesh hands, using mechanical hands may help then. Although usage may vary, as the dataset is not trained on mechanical hands.

• topless male in combination with cyborg helps generating him where his cyborg torso is shown. The dataset this LoRA was trained on doesn't actually have a lot of image with him like this, and the ones that are in the dataset only have like upper torso and up. So the cyborg part is to make sure Genos isn't generated with a flesh torso. Although not totally lore accurate it looked good enough in my testing.

• sleeveless, if you want to have his sleeveless look from the clothing he wears in the show you can add this to your prompt. Again same as the mechanical arms, can also work by putting it in the negative prompt for e.g. long sleeved clothing.

• The following prompts will generate him with clothing he is known to wear: blue shirt, white shirt, black pants, hood, hoodie, grey shirt, jeans & boots.

• The following prompts will generate him with his cyborg body glowing: glowing, glowing eyes & glowing hand. You can use them together or just pick the one(s) you want.

  • glowing hand might be a hit or miss, because hands and stable diffusion... It will try to attempt the hand beam pose.

• fighting stance, the dataset this LoRA has been trained on has Genos in various fighting poses. This prompt generates him in certain stances. Also including standing may force him in a standing fighting stance.

  • clenched hand may also help generate an image where is is clenching his hand in a fighting stance.

• floating hair may also sometimes work to generate him where his hair is blown back, like e.g. in the show when he was training with Saitama.

• It seems the LoRA is also flexible enough to genderbend Genos, if you're into that... It seems just prompting that Genos has breasts, does the trick. I've also uploaded some example images regarding this.

V2 is now out i hope u like the new version.
for V3 i need your help i need new fisting pictures, when u have good results with good anatomy pls post it anmd i will use it in v3 ^^.

城娘：Phoenix-cg

私服：Phoenix-pc

使用0.3强度需要对角色描述

Google Translate:

castle girl: Phoenix-cg

plain clothes: Phoenix-pc

Using 0.3 strength requires a description of the character

0.3, 0.3-0.6

Aloy from the game Horizon Zero Dawn.

This is my second Lora and it's much better than the first.

If you like share images!

I was planning to do a Cody Travers lora for a long time, then I realized that today (Apr 18) is his birthday. I still need to do some tests and I want to post a version with Mayor and Prisoner. But for now, enjoy the classic Cody from first Final Fight (my fave version 30 years later)

keyword is CodyTravers

works with Classic, white t-shirt, ripped shirt, shirtless... Although my database had a lot of his final fight look, t-shirt wants to disappear all the time. I need to test more, I just wanted to post on his birthday :P

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

STRIP CLUB

No descriptions needed, we both know why you're here...

Here's the way to the strip club!

This loRA is inspired by the style of the author 儒宅

https://www.pixiv.net/en/users/12539859

Please do not abuse copyright and respect the author

who else but quagmire!

0.8 is recommended

All images were generated using abyssorangemixs-2-Sfw

Hime Shiraki

Hime Shiraki (白木 陽芽, Shiraki Hime), known as Hime Shirasagi (白鷺 陽芽, Shirasagi Hime) at Liebe Girls Academy, is a highschool student who works part time at Cafe Liebe.

This character comes from Yuri Is My Job! (Japanese: 私の百合はお仕事です!, Hepburn: Watashi no Yuri wa Oshigoto Desu!) a Japanese yuri manga written and illustrated by Miman. It has been serialized in Ichijinsha's Comic Yuri Hime since November 2016. An anime television series adaptation premiered in April 2023.

Lora for nagisa from doaxvv. This was trained with 饭特稀_v0.8 which you can find on civit. The lora should work ok on similar models but it looks best on 饭特稀_v0.8. Trigger word is "nagisa".

Not trained on any particular costume.

Also recommend EasyNegative and 0.6-0.7 weight.

Chloe Mansfield クロエ・マンスフィールド | World's End Harem 終末のハーレム

https://www.pixiv.net/users/86166786

地点：(Night:1.7), Cyberpunk_CityView, Before Window,
动作：sitting at attention,sitting on couch,
服饰：hat,jacket,bare_shoulders,black_gloves,black_bikini,black_choker,bandeau,collarbone,underboob,navel,a short skirt,boots,belt,
角色：<lora:ChloeMansfield-V1:0.8>,<lora:aiueOkaStyle_v10:0.1>, <lora:saiminSeishidou:0.1>,<lora:odaNonStyleLORA_v2:0.2>, oda Non,
头发：blue_eyes,blonde_hair, center_parted_bangs,bare_forehead,
角色鎖定：
1 girl,Solo, Manga,27yo,mature female,Beautiful Finger,Beautiful long legs,Beautiful body,Beautiful Nose,Beautiful character design, perfect eyes, perfect face,
眼神表情：
looking at viewer,
畫質:
official art,extremely detailed CG unity 8k wallpaper, perfect lighting,Colorful, Bright_Front_face_Lighting,
(masterpiece:1.0),(best_quality:1.0), ultra high res,4K,ultra-detailed,
photography, 8K, HDR, highres, absurdres:1.2, Kodak portra 400, film grain, blurry background, bokeh:1.2, lens flare, (vibrant_color:1.2)
身材:
(Beautiful,large_Breasts:1.4), (beautiful_face:1.5),(narrow_waist),
Negative prompt: EasyNegative, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, Missing vagina, Blurry faces, Blank faces, bad face, Ugly, extra ear, amputee, missing hands, missing arms, missing legs, Extra fingers, 6 fingers, Extra feet, Missing nipples, ghost, futanari, Extra legs, Extra hands, panties,pants, (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), text, error, blurry, jpeg artifacts, cropped, normal quality, artist name, (worst quality, low quality:1.4),
Steps: 27, Sampler: Euler a, CFG scale: 7, Seed: 2009307287, Size: 512x768, Model hash: 3e9211917c, Model: yesmix_v16Original, Denoising strength: 0.5, Hires upscale: 2, Hires steps: 18, Hires upscaler: R-ESRGAN 4x+ Anime6B

The Juice is in the description below

Tips to use:

• Use LoRA weight between 0.6-0.8

• To summon her use the trigger Tags plus following tags: 1girl, horse ears, animal ears, hair intakes, horse girl, twintails, horse tail, brown hair, solo, tiara, epaulettes, large breasts, garter straps, very long hair, thighhighs, framed breasts, long sleeves, underbust, red eyes, skirt, blush, center frills, aiguillette, white skirt, bow, skin fang, ;d

We used one of Meina's model to generate the images.

If you have any questions feel free to ask!

Hope you guys like it! ❤️

You can ask for character LoRAs on our patreon page. Consider supporting us. 😘

Demo images use Danger Zoo ~ Johnsons Negative TI Collection ~

"modelname style" is the main trigger word
additional trigger words are listed to keep you close to the trained style/concepts.

This LORA is designed for used with other djz LORA's and/or on top of Mixed pairs of djz CKPT's or a Triple merge mix including other CKPT models. This is a specific method for merged CKPT models while running a stack of these Style LORA models on top. You may still use this alone and it is recommended to include the additional trigger words in your prompts.

DJZ Season 2.5

Demo images use Danger Zoo ~ Johnsons Negative TI Collection ~

"modelname style" is the main trigger word
additional trigger words are listed to keep you close to the trained style/concepts.

This LORA is designed for used with other djz LORA's and/or on top of Mixed pairs of djz CKPT's or a Triple merge mix including other CKPT models. This is a specific method for merged CKPT models while running a stack of these Style LORA models on top. You may still use this alone and it is recommended to include the additional trigger words in your prompts.

DJZ Season 2.5

Tsukuyo from Gintama

Trigger the look with Tsukuyo

hair between eyes works well in triggering her more casual look.

Kimono and fishnet thighhighs gets you her outfit.

Trained on AOM3

Demo images use Danger Zoo ~ Johnsons Negative TI Collection ~

"modelname style" is the main trigger word
additional trigger words are listed to keep you close to the trained style/concepts.

This LORA is designed for used with other djz LORA's and/or on top of Mixed pairs of djz CKPT's or a Triple merge mix including other CKPT models. This is a specific method for merged CKPT models while running a stack of these Style LORA models on top. You may still use this alone and it is recommended to include the additional trigger words in your prompts.

DJZ Season 2.5

Demo images use Danger Zoo ~ Johnsons Negative TI Collection ~

"modelname style" is the main trigger word
additional trigger words are listed to keep you close to the trained style/concepts.

This LORA is designed for used with other djz LORA's and/or on top of Mixed pairs of djz CKPT's or a Triple merge mix including other CKPT models. This is a specific method for merged CKPT models while running a stack of these Style LORA models on top. You may still use this alone and it is recommended to include the additional trigger words in your prompts.

DJZ Season 2.5

Demo images use Danger Zoo ~ Johnsons Negative TI Collection ~

"modelname style" is the main trigger word
additional trigger words are listed to keep you close to the trained style/concepts.

This LORA is designed for used with other djz LORA's and/or on top of Mixed pairs of djz CKPT's or a Triple merge mix including other CKPT models. This is a specific method for merged CKPT models while running a stack of these Style LORA models on top. You may still use this alone and it is recommended to include the additional trigger words in your prompts.

DJZ Season 2.5

Remember to use the trigger word choiyewon. Use LORA weight 1 and Hires fix with latent bicubic antialiased, 20 steps, 0.4-0.6 denoising strength, and 1.25x upscale. I've used the DDetailer extension for pretty much every example picture.

Version 2: You can use V2 with the same settings (as well as DDetailer to help) but you can also try mixing the two versions, ie, use "<lora:arinOTW_v2:0.5> <lora:arinOTW_v1:0.5>"

anything v4.5 checkpoint, anything v4.0 vae, clip skip 2

haven't test much. let me know what you think !

support the artist: https://twitter.com/09ra_19ra

Model that enhances images from space, have a nice look in most models.

In my opinion gives also a fantasy look to the images.

Try it, have fun.

Igrene from Fire Emblem The Binding Blade, trained on animefull

Works best between 0.6-0.9 weight

It gives various devices a demon maw.

Perpetual coffee request: I signed up for Buy-Me-a-Coffee and Ko-fi, mainly because I thought it would be hilarious if someone as artistically untalented as myself received a coin for something creative. It's already happened, and it made my day. AI made the impossible possible :D. If you'd like to add to the amusement, please consider buying me more coffee through the following links:

https://www.buymeacoffee.com/konyconiN

https://ko-fi.com/konyconi7992

Also, don't hesitate to send me suggestions. I tend to alternate between creating what I want and what I'm asked for. Rest assured, nothing goes unnoticed.

这个lora可以十分笨拙的把你心爱的女孩变成足盒，但因为训练集非常稀少（入选10张），它需要一些tag的指引才能发挥作用，而且不可避免的会使画面下降。

请使用：PrimeMix | Stable Diffusion Checkpoint | Civitai

或者：YesMix | Stable Diffusion Checkpoint | Civitai

直接上：AnyLoRA - Checkpoint | Stable Diffusion Checkpoint | Civitai

最好不要有任何描述腿部和身体的tag.

我推荐以：the wall photo have a girl portrait,in photo,pov,feetbox,only head,box,soles

This lora can very clumsily turn your beloved girl into a footbox, but because the training set is very scarce, it needs some tag guidance to work, and inevitably the picture will decrease. 

This loRA copies the style of the author 御坂12003@土曜日東ツ38(Misaka12003).

https://www.pixiv.net/en/users/2864095

Please do not abuse copyright.

Comparisons of samplers and forces of loRA are in art section.

German youtuber and streamer Anni The Duck.

Use lower strength (0.7 and below)

Since this is my first training I would appreciate some tips on improving.
Used 40 images with 50 steps (2000 steps total). At higher strengths it still might look overcooked and it can loose likeness on further distance.

Providing coffees or commissions start here!
Also check out my twitter~

Winner from twitter poll: Crystal!
Hi yalls~~ It took me a while to hone in on this one. It still has a lot of trouble with pixelation at the lower eyelids? I'll be looking at releasing another version of this model in due time. Stay tuned!

Trained on NAI with 84 pictures (mostly fanart).
Best result with weight between : 0.8-1.

Prompt activator : 4 outfits to prompt
Stage1: fdstg1, scar, cleavage, large breasts, long hair, pink hair, red coat, smile, white pants, thigh boots, high heels
Stage2: fdstg2, pirate hat, sleeveless
Stage3 - doesn't work // gacha : fdstg3, blue jacket over shoulders, epaulettes
Stormy : fdstm, eyepatch, scar, bare shoulders, white shirt, corset, cleavage, large breasts, ponytail, pink hair, belt, fingerless black gloves, smile, white pants, thigh boots, high heels
Negative : red coat

Commissioned by ohn0es on fiverr

Reviews are really appreciated, i love to see the community use my work, that's why I share it.
If you like my work, you can tip me here.

Got a specific request ? I'm open for commission on my fiverr gig *! If you provide enough data, OCs are accepted

*Any paid request will not be presented on this page, unless you give me permission to.

This is my second trained model. I haven't figured out how to train LORA yet. Basically everything is using the default preset. So I'm still a novice. I found that if using the 1.5 base model for training, the resulting LORA doesn't seem to match any CHECKPOINT, especially the commonly used CHILLOUTMIX. The characters become very childish. So I trained this LORA with CHILLOUTMIX. The effect of using it to generate images is better. Personally, I prefer more realistic characters instead of cosmetic faces. 

It may be because the pictures I used for training were 512*512. When generating images with 512*512, the images will be slightly distorted. Using 768*768 will improve a lot. To generate a half-length image, at least 768*1024 is required to have a better effect

weight 0.7 is best.

shaded face prompt is really important. If you want her face to be shaded.

gotou1, gotou hitori, solo, skirt, pink jacket, track jacket, bangs, hair between eyes, long sleeves, medium breasts
+ shaded face

gotou1, gotou hitori, solo, bangs, hair between eyes, short sleeves, maid, maid headdress, medium breasts
+ shaded face

gotou1, gotou hitori, solo, bangs, hair between eyes, short sleeves, school uniform, skirt, shirt, sailor collar, medium breasts
+ shaded face

gotou1, gotou hitori, solo, bangs, hair between eyes, playboy bunny, rabbit ears, pantyhose, medium breasts
+ shaded face

This is the 100th LoRA I've uploaded to civitai.

I only accept character requests on my discord.

caught

trained on solo and... coop scenes, but solo works best

using caught gives the character a surprised and guilty expression, works pretty well

combine with my female_masturbation model for bonus points

Kionant - Artist Style 🎨

This Lora is trained on the cute artworks of Kionant (NFSW Artist), mostly known for drawing monster girls, and pokemons, in you're interested make sure to check his work or Twitter or Instagram .

Usage Tips :

• regular use is at the weight of 1 to 0.9 , don't be afraid to experiment with higher values.

• Experiment with adding kionant to the prompt to feather anchor the style.

• This LoRA will try to make make the proportions super stylized, if you dont want it to have a strong affect on the shapes use with weight lower then 0.75.

Known issues :

• Colored skin, solved by adding colored skin, pokemon \(creature\)to the Negative prompt.

• elf ears, solved by adding pointy ears to the Negative prompt.

Samples Settings :

• Checkpoints : AOM2, AOM3, Grapefruit Mix

• Steps : 20 to 25

• Sampler: DPM++ 2M Karras

• CFG : 7

Weight Array:

Consider posting ur creation, I love to what u make with it 🤩

Tron Bonne Mega Man Legends

IMPORTANT!!!
if the file extension is .acc please change it to .safetensors maybe it is a page error

Hope you like it!

Recommended weight 0.7

Recommended prompting

green eyes, brown hair, gloves, earrings, hairband, jewelry, cropped jacket, pantyhose, pink gloves, hair pulled back

气抖冷 c站居然看不到阿波尼亚的LOAR 所以自己训练了一个 新手作品 略有瑕疵 尽请见谅

By request of a civitai's user.

This LoRA has been trained after carefully selecting 19 images, mostly from screenshots of Battlestar Galactica tv-shows. 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 epicDiffusion and RealisticVision checkpoints, although I think it will be just fine in other SD 1.5 models as well.

CFG scale should be between 4-7.

LoRA weights should be between 0.7-0.9

This is an attempt "LoRa" of Chamber from Valorant.

it was trained on anything v4.5, but it works on different models with minor tweaks here and there.

Use this initial prompt to try to give the character more strength:
"1boy, mature male, (chamber \(valorant\):0.85), glasses, <lora:CHMB:1.7>," with EasyNegative and BadHandv4 or the alternative Negative in the images.

It is highly recommended that you emphasize the clothing and background tags more by using parentheses and double punctuation.

It's a early version, it will be updated in the future if i can get more cinematic style chamber faces, because the LORA was trained with the chamber presentation trailer

Tinker bell is a well-known Disney character from Peter Pan. To create the Lora, a meticulously selected and refined DataSet was used to achieve a versatile and flexible model. So that it adapts perfectly to styles and models. It also works in 3D. Use the keyword “(TinkerWaifu:1)” for greater precision, and I also recommend using the keywords “(single hair bun), (strapless green dress), (fairy wings) to invoke her characteristic outfits. If you want to change her outfit, I recommend lowering the weight of the Lora between “0.8 - 0.5” for greater flexibility, try it out and have fun. Remember to leave a Review with an image if you liked it and comment!! HAVE FUN :D

Don’t forget to check out my other Disney Series Loras on my profile. Currently Rapunzel, Snow White, Merida, Cinderella, Pocahontas, Fa Mulan, Elsa, Anna, Tiana and Alice in Wonderland !! From the Princess Series ;D” Also u have The Bonus Series with: Nani Pelekai, Lifeguard (lilo & stitch).

Low weights are recommended for this model

Claudia Levantine クラウディア | Kuroinu 黒獣

https://www.pixiv.net/users/86166786

Thank to https://civitai.com/models/18396/claudia-levantine-kuroinu

I just want it to be perfect....

地点：(Night:1.7), in the Castle,medieval ,
动作：sitting at attention,sitting on couch,
服饰：(helmet:1.3),(armor),(breastplate),(skirt_armor),(silver_BRA),a helmet on her head,armored_boots,
角色：<lora:Claudia-V2:0.7>,<lora:claudiaLevantine_10:0.3>,<lora:aiueOkaStyle_v10:0.1>, <lora:saiminSeishidou:0.1>,<lora:olgaDiscordiaLORA_v6:0.1>,Queen,<lora:odaNonStyleLORA_v2:0.2>, oda Non,<lora:Claudia-V1:0.3>,
头发：(Blue_eyes),(blonde_short_hair), (hair_Bun),(Bangs),
角色鎖定：
1 girl,Solo, Manga,20yo,mature female,Beautiful Finger,Beautiful long legs,Beautiful body,Beautiful Nose,Beautiful character design, perfect eyes, perfect face,
眼神表情：
looking at viewer, (innocent_big_eyes:1.0),(expressionless:1.3),
畫質:
official art,extremely detailed CG unity 8k wallpaper, perfect lighting,Colorful, Bright_Front_face_Lighting,
(masterpiece:1.0),(best_quality:1.0), ultra high res,4K,ultra-detailed,
photography, 8K, HDR, highres, absurdres:1.2, Kodak portra 400, film grain, blurry background, bokeh:1.2, lens flare, (vibrant_color:1.2)
身材:
(Beautiful,large_Breasts:1.4), (beautiful_face:1.5),(narrow_waist),
Negative prompt: EasyNegative, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, Missing vagina, Blurry faces, Blank faces, bad face, Ugly, extra ear, amputee, missing hands, missing arms, missing legs, Extra fingers, 6 fingers, Extra feet, Missing nipples, ghost, futanari, Extra legs, Extra hands, panties,pants, (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), text, error, blurry, jpeg artifacts, cropped, normal quality, artist name, (worst quality, low quality:1.4),
Steps: 27, Sampler: Euler a, CFG scale: 7, Seed: 982853622, Size: 512x768, Model hash: 3e9211917c, Model: yesmix_v16Original, Denoising strength: 0.5, Hires upscale: 2, Hires steps: 18, Hires upscaler: R-ESRGAN 4x+ Anime6B

Please post in comments the photos that you manage to generate so we can all see them.

Trained on NAI with 80 pictures (mostly 3d fanart). Versatile lora (anime, semi-realistic, 3D, realistic)
Best result with weight between : 0.8-1.

Prompt activator : kasumi \(doa\)
Other prompt : brown hair, long hair, ponytail, bare shoulders, arm guards, pelvic curtain

Torn clothes prompt : torn clothes, dirty, ripped clothes

For full body pictures, sometimes you will need to inpaint the face because it can be distorted.

Commissioned by ocelot1300 on fiverr

Reviews are really appreciated, i love to see the community use my work, that's why I share it.
If you like my work, you can tip me here.

Got a specific request ? I'm open for commission on my fiverr gig *! If you provide enough data, OCs are accepted

*Any paid request will not be presented on this page, unless you give me permission to.

Nothing to add, the model is perfect.

Adds a cute bulge to all the images, seems to be doing pretty well out of 30 runs - they all came out fine.

The 12 epch version also works well, both are left, the 20 version has more training

bulge,<lora:bulg-20:1>----->Activacion

bulge (51)
solo (48)
navel (46)
underwear (43)
blush (42)
breasts (41)
panties (40)
1girl (37)
looking at viewer (36)
futanari (33)
male focus (31)
long hair (29)
large breasts (28)
thighhighs (28)
1boy (28)
smile (27)
otoko no ko (27)
penis (24)
blonde hair (22)
nipples (21)
standing (19)
open mouth (19)
bangs (18)
short hair (18)
erection (18)
testicles (18)
black hair (17)
blue eyes (17)
crossdressing (17)
covered nipples (16)
swimsuit (15)
gloves (15)
bare shoulders (14)
simple background (14)
bikini (13)
jewelry (13)
collarbone (12)
holding (12)
elbow gloves (12)
outdoors (11)
english text (11)
dark skin (10)
sweat (10)
multiple boys (10)
black thighhighs (10)
lips (10)
full body (10)
thighs (10)
earrings (10)
cleavage (9)
pubic hair (9)
yellow eyes (9)
erection under clothes (9)
2boys (9)
spread legs (9)
high heels (9)
dress (9)
tan (9)
skirt (9)
shiny (9)
nail polish (9)
purple eyes (8)
choker (8)
white panties (8)
female pubic hair (8)
sitting (8)
day (8)
solo focus (8)
hat (8)
bow (8)
animal ears (8)
white background (8)
striped (8)
thick thighs (8)
red eyes (7)

Please consider supporting me so I can continue to make more Models / LoRA's for the community! --> Buy Me A Coffee

In my pursuit of training a new Furry model I've begun to produce new loRA's, in preparation for creating a more varied, cohesive database from which to work off of.

This model handles both Feral and Anthro variants rather well. Also, the tags might be a bit strong on this one, so you might want to turn down the strength of the LoRA depending on what you're going for. Using the weights at 0.4 seems to work best for more varied results.

This LoRA was trained using Furry Vixens 2.0, and works exceptionally well with it! The images were generated using clip skip 2.

Image resolution for training was 512x786

You can use these tags when generating images:

Eevee, Eevee tail, Eevee ears, brown eyes, brown fur, autumn, forest, trees

Though obviously all other tags work well too! These might get you some more accurate Eevees though, depending on the rest of your prompt.

And be sure to use these settings!

Steps: 32

Sampler: DPM++ 2M Karras

CFG scale: 7

Seed: 3016242806

Size: 384x512

Model: Furry Vixens v2

Denoising strength: 0.6

Clip skip: 2

Hires upscale: 2

Hires upscaler: R-ESRGAN 4x+ Anime6B

Positive Prompt

Eevee, Eevee tail, Eevee ears, pawpads, smile, open mouth, sitting, looking at viewer, autumn, forest, trees, <lora:Eevee V1:0.7>

Negative prompt

(nsfw:1.4), (worst quality:1.4), (low quality:1.4)

This LoRa is made with the purpose of recreating the style of Zero no Tsukaima, it can still be improved more, and I hope to update it soon.

Thai Khon Hanuman

My atempt at creating a lora for Satanichia. Works suprisingly well. With her original clothes as well as some other.

Works well with weights 0.8-0.6. Of course lower weights work better when you want to achieve pics with her in something more casual.
Trained on anything v4.5. But it works also fine on orangemix models.

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

I'm open to Character + Style requests!

Weight: 0.7 - 0.8

Trigger + Appearance: fiona frost, short white hair, purple eyes, jewelry, earrings

She's usually wearing a blue coat or a black turtleneck with black pants and black gloves.

Nel from Fire Emblem Engage, trained on animefull

Recommended to use weight of 0.7-0.9

Due to the small amount of art available, I used the "aug_flip" setting to train this lora, as a result some things may show up on the wrong side, this will only affect symmetry, other than that, this lora should be up to our normal standards.

Trigger word : isa

V2

here : -> isa waiting for you :v <-

Use this wisely. I'm not responsible for anything you create with this LoRA.

Trigger word : dita

V2

here : -> halo kata mbak dita <-

Use this wisely. I'm not responsible for anything you create with this LoRA.

Trigger word : jeongyeon

V2

here : -> Short hair + tsundere <-

Use this wisely. I'm not responsible for anything you create with this LoRA.

これをまともに扱うの難しいかも

Sharla from Xenoblade Chronicles, this is the final of the 3 party member girls. Seems to produce decent results, sometimes it puts here mole in random places or adds multiple moles, so you might need to do a little inpainting.

Trigger word is 'sharla', if you want her regular outfit add: 'crop top, vest, white shorts, thigh boots'

Trained on NAI anime, 27 images

After scouring through hundreds of models and trying to generate a Terminator Endoskeleton it became obvious to me that very little of Terminator exists in any model's training data. So, using Hollowstraberry's tutorial & colab I trained this LORA on about 80 HD images at about 4000 steps on Experience 8.0. It does closeups the best, but full body shots are still quite good, and seems quite flexible with styles and different models.

I recommend a weight of 0.8 - 0.9

Adding "holding a gun" or "holding a futuristic gun shooting lasers" without quotations should produce the signature futuristic rifle the Terminators usually use in the future.

I recommend 512x512 or 768, LORA was trained at 768 resolution so does well with higher resolution. It seems to handle high res fix really well and produces amazing quality with it.

This is my second LORA and I don't really know what I'm doing outside of the tutorial and colab, I have a AMD 7900xt so I have to use the DirectML version of A1111 whichi doesn't currently work for local training for me. If there's any tips you might have, please let me know.

Please rate & comment and share your generations & let me know how I'm doing!

Madison Kate AKA Madikat

Sample images made using RealisticVision model.

My best results were keeping CFG between 3 and 4, steps around 20, and LoRA weight around 0.65

Lower the LoRA weight if using in conjunction with other LoRA's; inpaint face afterwards ('only masked' option) with higher weight for most accurate results.

The newest apparation in the Black Bullet roster : Kohina Hiruko

Use at 0.8~1.0 weights

The trigger word is "bbkohina"

Most of the dataset is outside and it will be difficult to trigger her classic dress in an inside environnement.

It was trained with mutilple tags regarding her dual wielding katana unfortunatly it's 80% miss. try at your own sanity.

Tested on :

• Greymix

• MidnightMelt

• AOM3

• Kotosmix

• Sweetmix

• Henssaku

• Mandarine

If you have any questions feel free to ask.

Kaïros#4463

Enjoy ♥

Louisa Richter ルイーザ・リヒター | Seika Jogakuin 聖華女学院

https://www.pixiv.net/users/86166786

地点：(Night:1.7), Cyberpunk_CityView, Before Window,
动作：sitting at attention,sitting on the couch,
服饰：school_uniform,White collared shirt, pleated skirt, red_bowtie,
角色：<lora:LouisaRichter-V1:0.7>,<lora:aiueOkaStyle_v10:0.1>, <lora:saiminSeishidou:0.1>,
头发：blonde hair ,green eyes,bangs,
角色鎖定：
1 girl,Solo, Manga,20yo,mature female,Beautiful Finger,Beautiful long legs,Beautiful body,Beautiful Nose,Beautiful character design, perfect eyes, perfect face,
眼神表情：
looking at viewer, (innocent_big_eyes:1.0),(expressionless:1.3),
畫質:
official art,extremely detailed CG unity 8k wallpaper, perfect lighting,Colorful, Bright_Front_face_Lighting,
(masterpiece:1.0),(best_quality:1.0), ultra high res,4K,ultra-detailed,
photography, 8K, HDR, highres, absurdres:1.2, Kodak portra 400, film grain, blurry background, bokeh:1.2, lens flare, (vibrant_color:1.2)
身材:
(Beautiful,large_Breasts:1.4), (beautiful_face:1.5),(narrow_waist),
Negative prompt: EasyNegative, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, jpeg artifacts, signature, watermark, username, blurry, out of focus, censorship, Missing vagina, Blurry faces, Blank faces, bad face, Ugly, extra ear, amputee, missing hands, missing arms, missing legs, Extra fingers, 6 fingers, Extra feet, Missing nipples, ghost, futanari, Extra legs, Extra hands, panties,pants, (painting by bad-artist-anime:0.9), (painting by bad-artist:0.9), text, error, blurry, jpeg artifacts, cropped, normal quality, artist name, (worst quality, low quality:1.4),
Steps: 27, Sampler: Euler a, CFG scale: 7, Seed: 2067574495, Size: 512x768, Model hash: 3e9211917c, Model: yesmix_v16Original, Denoising strength: 0.5, Hires upscale: 2, Hires steps: 18, Hires upscaler: R-ESRGAN 4x+ Anime6B

As requested. Works well enough, I think.

Trigger word is Throné.

Weight around ~0.7.

Looks mediocre on low resolution.

Use "purple dress". Also "hair over one eye" if you want.

Trained on NAI.

(As Requested)
Trained on 51 photos, work best on DPM++ SDE Karras sampling steps 20-30 CFG Scale 6 on ChillOutMix. (haven't tried on other models). weight 1 is recommended for better results.
lightweight LoRA, works on any generator (easydiffusion,stablediffusion etc). share your results.

Thank you for downloading.please leave a rating and comment if you like my models. also leave a comment on what model would you like to see next. enjoy

update 20230419

new locon version released,recommend weight 0.6~0.8

V1.0

这次降低了步数，使得模型没有以前那么硬，可以适当调高权重以获得比较明显的画师风格，推荐权重0.8，模型除了增加风格外好像还能提升黑丝的质感大家可以亲自尝试一下

This time, the number of steps has been reduced, so that the model is not as hard as before, you can properly adjust the weight to get a more obvious painter's style, the recommended weight is 0.8, the model in addition to increasing the style seems to enhance the texture of legwears, enjoy the model and try it!

A SD1.5 based LoRA based on Mitsurugi from Soulcalibur VI. Trained at a resolution of 576 x 576, so use that as your base resolution and upscale with Hires fix if needed. The optimal strength for this LoRA ranges between 0.8-1.0

v1.1 changes the tagging of the images to allow for easier outfit swapping. See the 3rd and 4th sample image prompts for examples.

Velma played by Linda Cardellini in the movie Scooby-Doo (2002).

Please provide constructive criticism.

You can support me here to be part of the society (https://ko-fi.com/djo226). Supporters may request model.

Prompt : Velma , high quality photography, natural lighting, canon eos 5d mark ii, photorealistic, 8k, photo by Annie Leibovitz, detailed skin, ultra realistic, bokeh, sharp features, centered in frame, facing camera, symmetrical face, ideal human, 90mm lens, f16, photography, ultra details, photo <lora:Velma:0.8>

Negative prompt: (deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands and fingers:1.4), (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, disconnected limbs, mutation, mutated, ugly, disgusting, amputation

The tagger word is "<niji-se-3d-bikini>".

We use https://civitai.com/models/7371/rev-animated as our base model. For convenience, we provide our prompts for easy beginnings. We train with images of swimsuit (boys and girls), and hope you enjoy the summer via this model!

It would be very appreciated if you could share with me your interesting results. For any problem, feel free to ask in this post. Note that this version is inclined to close eye and performs better on girls, we will add more training data and fix this in the next version.

Adds a flaccid, non-functional micro penis to the model.

micropenis,<lora:small-10:1>---->Activacion

penis (58) ----> micropenis
testicles (57)
solo (56)
navel (50)
looking at viewer (49)
nipples (45)
flaccid (39)
1boy (38)
blush (38)
short hair (36)
male focus (35)
uncensored (35)
foreskin (35)
breasts (35)
1girl (33)
nude (28)
otoko no ko (27)
futanari (27)
smile (26)
censored (24)
thighhighs (23)
open mouth (23)
simple background (20)
black hair (20)
long hair (20)
sweat (17)
erection (17)
standing (17)
choker (16)
cowboy shot (15)
jewelry (15)
large breasts (15)
bangs (15)
dark skin (14)
earrings (14)
makeup (14)
blue eyes (13)
white background (13)
thighs (13)
collarbone (13)
artist name (13)
pussy (12)
bar censor (12)
phimosis (12)
gloves (11)
outdoors (11)
blonde hair (11)
tail (11)
english text (11)
green eyes (11)
piercing (11)
pectorals (10)
thick thighs (10)
lipstick (10)
lips (10)
black thighhighs (10)
sitting (9)
muscular (9)
dark-skinned male (9)
monochrome (9)
animal ears (9)
hair between eyes (9)
underwear (9)
solo focus (8)
crossdressing (8)
parted lips (8)
ass (8)
small breasts (8)
nipple piercing (8)
small penis (8)
multicolored hair (8)
tongue (8)
spread legs (7)
closed mouth (7)
abs (7)

For Ai chat, lemon bot.

For me, wel-cum

챗붕이들 쓰셈

You can get an idea about the specifics of this LoRA by looking at example images.

The base model is Suzumehachi.

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

I'm open to Character + Style requests!

Weight: 0.7 - 0.8

Trigger + Appearance + Outfit: gfl-ak47, red beret, green crop top, sleeveless, black shorts, white gloves, cleavage, large breasts

Add blonde hair and blue eyes to the positive prompt if necessary.

Penélope Cruz - Spanish actress.

Please provide constructive criticism.

You can support me here to be part of the society (https://ko-fi.com/djo226). Supporters may request model.

Prompt : Penélope Cruz , high quality photography, natural lighting, canon eos 5d mark ii, photorealistic, 8k, photo by Annie Leibovitz, detailed skin, ultra realistic, bokeh, sharp features, centered in frame, facing camera, symmetrical face, ideal human, 90mm lens, f16, photography, ultra details, photo <lora:Penélope Cruz:0.7>

Negative prompt: (deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands and fingers:1.4), (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, disconnected limbs, mutation, mutated, ugly, disgusting, amputation

Steps: 80, Sampler: DPM++ 2M Karras, CFG scale: 12, Seed: 739720113, Face restoration: GFPGAN, Size: 768x512, Model hash: 81086e2b3f, Model: realisticVisionV20_v20

Face restoration is optional, may provide better result if not activated.

I think it speaks for itself, use with the tag 'xbv1' for slightly better results. Works well with dreamshaper, deliberate and analogMadness, but haven't had too much trouble in other ones I been playing with. It does lean towards the nsfw so you may want to put nsfw etc. in the negative prompt if that isn't what you're going for. Trained with 31 high-res images with no edits. Anyways... Enjoy!

Note:
Someone asked for the rest of the generation data in a comment on an image, so I replaced the images (which deleted the comment...oops :/ ) with the ones from the output folder, apparently moving them to another drive, messed with the png headers somehow? Idk whatever, it's fixed now :) apologies

This is my first time trainning model, welcome to give me feedback

Japanese actress 武田 玲奈

Legal & Risk

It is prohibited to use this model for commercial purposes and any scenarios of illegal acts and purposes.

This was trained using Realistic Vision 1.3

It is my first lora, so don't expect perfect results.

I'll keep improving to release something more accurate

这是一个测试

That's a start

Trained for 2000 steps

"#:0.04|highly detailed, sharp focus, intricate, smooth, elegant, 8 k, fantasy, cinematic lighting, cinematic, masterpiece, matte, photorealistic, 4 k, 8k, beautiful, volumetric lighting, dramatic lighting, detailed, realistic, intricate details, ultra realistic, high detail, centered, hyper detailed, 4k, 8K, hyperrealistic, hd:0.2"

https://github.com/ntc-ai/conceptmod

This prompt is the most popular prompt phrases from:
https://huggingface.co/datasets/Gustavosta/Stable-Diffusion-Prompts
by usage.

animation:

https://github.com/ntc-ai/conceptmod/blob/main/lora_anim.py

This does not include any people, artists or websites so use it however you want.

Trigger word : BLAshadow

You can also add : blinds, window, shadow

Recommended to use with epi_noiseoffset : https://civitai.com/models/13941/epinoiseoffset

https://www.aiartbot.com is a website I am currently operating that compares AI models with keywords. I hope it can provide some assistance for you.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

触发词：BLAshadow

你也可以选用：blinds, window, shadow

推荐与 epi_noiseoffset 搭配使用：https://civitai.com/models/13941/epinoiseoffset

https://www.aiartbot.com 这是我正在运营的比较AI模型与关键词的网站，希望它能对你有所帮助。

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

트리거워드 : BLAshadow

추천드리는 키워드 : blinds, window, shadow

epi_noiseoffset 와 함께 사용하시는걸 추천드립니다.https://civitai.com/models/13941/epinoiseoffset

https://www.aiartbot.com 제가 운영하고있는 모델와 키워드를 비교하는 사이트 입니다. 작은 도움이라도 되기를 바랍니다.

Bonus characters!! Stepping away from the Disney princess series, I bring you 2 characters that I think you’ll love!! Nani Pelekai and Lifeguard from Lilo & Stitch. I hope you enjoy it!!

To create the Lora, a meticulously selected and refined DataSet was used to achieve a versatile and flexible model. So that it adapts perfectly to styles and models. It also works in 3D. Use the keyword “(lifeguardWaifu:1)” for greater precision, and I also recommend using the keywords “(red one-piece swimsuit), to invoke her characteristic outfits. If you want to change her outfit, I recommend lowering the weight of the Lora between “0.8 - 0.5” for greater flexibility, try it out and have fun. Remember to leave a Review with an image if you liked it and comment!! HAVE FUN :D

Don’t forget to check out my other Disney Series Loras on my profile. Currently Rapunzel, Snow White, Merida, Cinderella, Pocahontas, Fa Mulan, Elsa, Anna and Alice in Wonderland!! ;D”

If you like my LoRAs please consider to support me by buying me a coffee here !

And if you want to see my work diary and notice post of next LoRAs, click here

Model : chilloutmix_NiPrunedFp32Fix

Vae : vae-ft-mse-840000-ema-pruned

Please do not post NSFW or sexual pictures.

Bonus characters!! Stepping away from the Disney princess series, I bring you 2 characters that I think you’ll love!! Nani Pelekai and Lifeguard from Lilo & Stitch. I hope you enjoy it!!

To create the Lora, a meticulously selected and refined DataSet was used to achieve a versatile and flexible model. So that it adapts perfectly to styles and models. It also works in 3D. Use the keyword “(NaniWaifu:1)” for greater precision, and I also recommend using the keywords “(striped bikini), (swimsuit, one-piece swimsuit) and (crop top, shorts, hair flower)” to invoke her characteristic outfits. If you want to change her outfit, I recommend lowering the weight of the Lora between “0.8 - 0.5” for greater flexibility, try it out and have fun. Remember to leave a Review with an image if you liked it and comment!! HAVE FUN :D

Don’t forget to check out my other Disney Series Loras on my profile. Currently Rapunzel, Snow White, Merida, Cinderella, Pocahontas, Fa Mulan, Elsa, Anna and Alice in Wonderland!! ;D”

LORA ของแป้งสามป๋องซาว ที่ได้รับการรับรองจากเจ้าตัวแล้วครับ กรุณาอย่าใช้เจนรูป 18+ และเผยแพร่ รับผิดชอบตัวเองกันด้วยนะครับ

This is trained LORA of Pang3pong Thai game steamer and influencer has approved by herself.

I generally do not support the use of this model to generate any adult NSFW 18+ content. When using this model, users should abide by local laws and regulations and shall not violate any other's right to reputation, privacy or portrait.

trained with the 1.5 base model, It should work well with any photorealistic models, You'll get best results with a LORA weight 0.6-0.8, Steps: 25-40, Sampler: DPM++ 2M Karras, CFG scale: 7-8

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

I'm open to Character + Style requests!

Weight: 0.7 - 0.8

Trigger + Outfit: hilda \(beelzebub\), maid uniform, black elbow gloves, choker, black footwear, knee boots

Add blonde hair and green eyes to the positive prompt if necessary.

Dataset comprises of 31 fan art images and 28 screen caps.

This is a LoRA for generating images of Persona 5 character Yoshizawa Kasumi.

Making models takes a lot of time and hard work. If you like this LoRA, please click the like button!

Recommended weights:

• Recommended weights: 0.5~0.7

  (Note: If the weight is below 0.4, the LoRA may not work effectively.)

這是一個用於生成女神異聞錄5角色芳澤霞的圖像的LoRA

製作模型需要花費很多的時間和精力。如果您喜歡這個LoRA，請幫忙按一個“讚”！

建議權重：

• 建議權重：0.5〜0.7

  （注意：如果權重低於0.4，則LoRA可能無法生效。）

用的真人模型，权重0.6左右最好，喜欢的请来小红书多多支持，小红书唯一账号122766380，不定时更新，后续系列lora会慢慢上传~

See image prompts info for checkpoints. Results will vary depending on the model.

Remember to use the trigger word jeonsomi. You can use LORA weight 1 and Hires fix with latent bicubic antialiased, 20 steps, 0.4-0.6 denoising strength, and 1.25x upscale. Inpainting can help sometimes. Might need to negative prompt "blonde hair, pink hair" if you don't want it since most of the training images were on light colored hair Somi.

See image prompt info for checkpoints. Results will vary depending on the model.

I recommend 0.5-0.7~ and 5 CFG for accurate looking kyary.
0.2-0.4~ for general beauty/cuteness.

Keyword is pamyurin, 1girl, fake eyelashes, long eyelashes, blush but leave it out if for general beauty/cuteness.

Trained on a mixture of aresmix and marsmix.

Note, this is my first time training a LoRA. I tried training on SD1.5 but the result came out bad with this same dataset.

my links: https://rentry.org/bericbone

请不要使用该模型用于商业用途。

Please do not use this model for commercial purposes.

角色原型来自游戏：Blue Archive

**以下内容使用机器翻译 The following uses machine translation（To English)**

角色原型来自游戏：Blue Archive

The character prototype comes from the game: Blue Archive

用合适的tag去描述符合角色原本的年龄、性别特征可能获得更好的成果。

Using appropriate tags to describe the original age and gender characteristics of the character may obtain better results.

当前上传的LoRa可能更适用于AnythingV4.5 之后可能会上传其他版本，也可能不会。

The currently uploaded LoRa may be more suitable for AnythingV4.5. Other versions may or may not be uploaded.

这种LoRa应该没有提示词，请直接添加 <lora:isamikaedeV010:1>使用。

This LoRa should have no prompt, please add <lora:isamikaedeV010:1>to tags for use it.

推荐使用的Tag/Recommended tag：

• 通用/General

  <lora:isamikaedeV010:0.8>, yellow_eyes, large_breasts, double_bun, animal_ears, hair_bun, black_hair

• 尝试出现光环 Try to make a halo appear in the picture

  sakura_halo, halo

• 巫女服 /Miko clothes

  miko, nontraditional_miko, ribbon-trimmed_sleeves, japanese_clothes, sailor_collar, pleated_skirt, white_legwear, white_skirt,

• 泳装（水着/ Swimsuit

  one-piece_swimsuit, white_swimsuit

• 如果连续出现同一姿势的情况 请尝试在负面提示词中使用以下tag/If the same pose occurs in succession, please try to add the following tag in the Negative prompt:

  tachi-e, tachi-e_stand

trigger word : sn00pd0g

recomm. strength : 0.88 - 1.0 (works best),

do not use below 0.8, otherwise likeness will be lost

if you like my stuff pls consider supporting me on Kofi

If you don't have permission, please don't use a real face to make any work.

No business model, if there is any infringement, please contact me to delete

If you want, you can buy me a coke(patreon).

Thank you very much.

A LORA model trained on the Chris Evans actor.

To use place the words "chrisevans person" in the prompt.

Trained on 6150 steps

In my opinion, a very talented young actress. With a very unusual appearance.

Model trained on later photos of Lillis. Works best in the 0.7-0.8 range.

If you like our model, please buy me a coffee and supporting for future project.
https://ko-fi.com/aiphoto

After download this model, try to share with us about your finding.

Please use ChilloutMix and based on SD1.5

• 768,768 image

• Use DPM++ 2M Karras or DPM++ SDE Karras

• Use 40 sampling steps

• Strength: 0.8-1.0

这是一个水彩手绘风格的场景LoRA

底模推荐MeinaMix_meinaV9

VAE推荐autoencoder_fix_kl-f8-trinart_characters

权重推荐0.5-0.75，如与其他lora搭配请自行判断。

为爱发电

Works best at 0.8

Chevrolet 1970 Chevelle SS trained with 768x768 images. Classic "Supernatural" car!

LoRA strength: 0.6 - 0.8, 0.65 works best on revAnimated

All images were generated using abyssorangemixs-2-Sfw

New:

◍合并版

城娘：Tartarus-cg

私服：Tartarus-pc

castle girl: Tartarus-cg

plain clothes: Tartarus-pc

0.4-0.6

Old：

◍◍◍两件衣服单独在两个lora模型里◍◍◍

这个角色我不知道怎么了，我练了三次都没把两件衣服练进一个lora，另一件写触发词画出来也是上一件，先分开吧，我再去试试能不能塞一起。

城娘：Tartarus

私服：Tartarus-pc

Google Translate:

◍◍◍Two clothes are in two lora models separately◍◍◍

I don’t know the reason for this character. I trained three times but failed to train two clothes into one lora file. I wrote the prompt words for another clothes, but the AI drew the same clothes as before. Divide it into two loras, and I'll try to see if I can put them together.

castle girl: Tartarus

plain clothes: Tartarus-pc

第五人格3D风格大模型v1版本和Lora模型v3版本同时制作完成啦～

大模型需要和Lora模型v3版本同时使用，只使用大模型将没有效果。

推荐设置：

大模型：IdentityV_V1

Lora：identityvlora_v3

LoRA强度：0.3-0.4

采样器：DPM++ 2M Karras

采样步数：20

CFG：6-8

推荐尺寸：768*512

推荐使用高清修复

如果大模型不使用IdentityV_V1，推荐使用Revanimated，权重为0.8。

See image prompts info for checkpoints. Results will vary depending on the model.

Ayaka Hizuki from Memories Off 1st

V1.0

Recommend weight: 0.6-0.9

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

2023.4.18 update

V2.0

The art style now more closely resembles the original game.

The probability of generating wings with errors has decreased now.

Now you can get her wearing the school uniform without having to input any prompts.

If you want her to wear everyday clothes, please input "townwear" and"dress" as the prompt word.

If you want her to wear other clothes, please strengthen the weight of the prompt words, such as (((Nurse wearing))).

Recommend weight: 0.9

Recommend Checkpoint: final-prune, anything, mixpro or other anime checkpoint

Recommend negative prompt: own hands together

This is the Alpha test version, I trained it for several weeks and did not train the LoRA model to my satisfaction.

You may encounter the following problems for which I do not have an effective solution:

dual anus, pus piercing instead or clit piercing, no clit, no or ugly pus, nip piercing without positive prompt, outfitted,

example positive prompt: <lora:clitpiercing_test:0.65>, clitoris piercing, clitoris ring, clitoris,

Added Conv2D to improve quality. Increase the file size by half of DIM.

Training Data v.1.0

• 214 image 15 repeats 3 epochs

• Powershell ISE run script kohya-ss

• module = Networks.LoRA

• cosine_with_restarts 3

• 8 DIM / 4 Alpha

• Conv2D 4DIM / 4 Alpha

• U-net 1e-4

• ???

768*1080

步数30

CFG Scale 8

可以使用高清修复

20步

重绘0.3

算法 R-ESRGAN 4x+Anime6B

放大倍率2-2.5倍

模型建议使用yden这个模型，因为我是基于这个模型炼的lora，用其他模型出图效果会不大好

https://civitai.com/models/18616/yden

LoRA trained on images for large insertion with multiple positions.

You can use it with also with my recently posted BLACKED LoRA.

You can modify the weight (0.0-2.0) to increase the effect.

Saito Misaki, heroine of adult comic: "Yume ka Utsutsu ka" (Izumi, Reizei)

Legacy LORA Krekkov Style

This Lora is made up in the style of Krekkov, preserving for the community.

This is not the same LORA as:

https://civitai.com/models/35957/krekkov-style-lora

(But check theirs out too)

Prompting Info

• Really likes doing Androgynous characters, females with extra parts & large breasts / muscles and small feminine & sexualized guys, etc.

• Textual Inversions like easynegative & bad-hands-5 are your friend, anything that fixes bad anatomy.

Recommended Negative Prompt:

(worst quality, low quality:1.4), easynegative, bad anatomy, (text)

Credits

Artist is Krekkov

• https://www.pixiv.net/en/users/69447461

Original LORA author was hussienrahy, maybe? Not sure - everything was nuked about this Lora.

Model of Madonna at the beginning of her career.

郑重声明：本模型禁止用于训练基于明星、公众人物肖像的风格模型训练。

Hanfu Tang Style Lora model, feel the Tang Dynasty style!

汉服唐风Lora模型，感受唐朝风采吧！

汉服总模型：https://civitai.com/models/15365/hanfu

汉服唐风文档: https://sleepy-oyster-204.notion.site/tang-style-c4a0af02d2a844d1b8a4406e282291b7

基础模型BaseModel: chilloutmix、Five Nuts Mixed、Basil mix、cardosAnime、meinamix、RealisticVision

Hanfu ALL model: https://civitai.com/models/15365/hanfu

Hanfu Tang Style Document: https://sleepy-oyster-204.notion.site/tang-style-c4a0af02d2a844d1b8a4406e282291b7

QQ交流群 149774926

微信交流群添加小助手 hanfuaigc

唐3.0 tang

☑️ 上衫 upper shan ☑️ 长上衫 long upper shan

☑️ orange/red/green/blue/pink/yellow/white+upper shan ☑️ 橙/红/绿/蓝/粉/黄/白 + 上衫

☑️ 淡蓝 light blue ☑️ 浅绿 light green ☑️ 浅红 light red ☑️ 青蓝 cyan blue ☑️ 浅紫色 light purple ☑️ 黑蓝 black with blue ☑️ 橙白 white with orange ☑️ 粉黑 pink with black

☑️ 齐胸破裙 chest po skirt ☑️ 齐胸褶裙 chest pleated skirt

☑️ white/red/green/blue/pink/blue/purple + chest pleated skirt ☑️ 白/红/绿/蓝/粉/蓝/紫 + 齐胸破裙/齐胸褶裙

☑️ 彩色 multicolour ☑️ 紫白 purple with white ☑️ 黄白 white with yellow ☑️ 天蓝 blue with white ☑️ 橙白 orange with white ☑️ 灰豆绿 green with white ☑️ 淡黄白 light orange with white ☑️ 粉白裙子 pink with white / white with pink

☑️ 系带 waistband

☑️ green/yello/red/pink/orange + waistband ☑️ 绿/黄/红/粉/橙 + 系带

☑️ 淡蓝 light blue ☑️ 绿白 green with white ☑️ 黄绿 green with yellow ☑️ 紫白 purple with white ☑️ 红白系带 red with white waistband ☑️ 红黑系带 red with black waistband ☑️ 黄白系带 yellow with white waistband

☑️ 披帛 pibo

☑️ white/orange/black/pink/green/red/purple + pibo ☑️白/橙/黑/粉/绿/红/紫 + 披帛

☑️ 浅绿披帛 light green pibo ☑️ 黄白披帛 white with yellow pibo

Thank Support Contribution 感谢以下人员的支持与贡献

• @liaoliaojun @明月明月 @南陌 @现代汉服研习小组 @善果心花 @WEN @科晓生 @林子安 @A++ 设计师雨晨

• 首页精修图作者：设计师雨晨 https://v.douyin.com/DFCxhyD/

Another request and a 200% beefcake maker. Can also be used at lower weights as a general beefer upper. The effect seems to quite precise at leaving the face intact, great for anyone wanting to buff a particular character.

How to use
While no trigger word is required I highly recommend using the tag "muscular female" otherwise little to no effect is seen. Other optional tags are anything muscle related such as "biceps", "triceps", "deltoids", "obliques", "abs", "midriff" etc.
Depending on how you prompt you can reach all the way into hyper territory.

Strength can be anything up to 1.0. NAI doesn't need as high values as AOM2.

Don't forget to review and upload your own images, it's always fun to see what you're creating!
Like my LoRAs and want more? Please consider a small tip towards my electricity bill and further training >.<

Misc
Trained on NAI at 512 res. Dataset of 100 images.
With the exception of the inpainted face in the "realistic" image all images are raw, slightly cherrypicked, txt2img gens, no post work done.

Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord

Training settings:

35 images, 25 steps, 10 epochs, you find many epochs (with test images using plot xy script) on my Patreon while supporting me.

Recommended settings:

• Make sure you are aware on the usage instructions of LORA

• Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)

• CFG Scale : 5 to 9

• LORA weight for txt2img: reccomended 0.3-0.7

• Keys: wearing luxury dress, hotel lobby, shiny dress

• Adviced Image Size: 512x768

This LORA isn't easy to use, i advice to start the prompt with "a woman wearing luxury dress, DRESSCOLOR dress, shiny dress" or something similar

If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians

Introduction

A LoRA model of Aegir (Spring Festival outfit) from Azur Lane. The trigger tag is ‘spring-aegir’. The 784mb VAEs (NovelAI, Anythingv3, Orangemix, Counterfeit) are recommended. 0.6 ~ 0.8 weights are good. The preview images are generated using 1.6 version (original) YesMix.

Cautions (Must Read)

• Some preview images are generated using hires fix. Thus, if you wanna generate the preview images, pls download the original preview images and analyze them using 'PNG INFO' in webui to get all generating information.

• The maximum of generating resolution should not exceed 1024 theoretically, otherwise extra/deformed head/limbs will be waiting for you :). Try to use 'hires fix' to prevent this problem and help to generate higher-resolution images.

• For 784mb VAEs (NovelAI, Anythingv3, Orangemix, Counterfeit) users, make sure add '--no-half-vae' in command line of webui to prevent generating black images. See voldy's log for the details of edit method. For high-performance GPU, make sure 'medvram' and 'lowvram' mode are disabled.

LORA for peeking out pose. The out layers have no contribution to the pose and have been removed from the file already (equivalent LBW: 1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0).

Sugggested weight range: 1~2 Default weight 1 should be good to start with. Some base models may require larger weights. Please refer to the XYZ plot to see how weight affects the pose.

Trigger words: peeking out upper body

Some useful scenery prompts: brick wall, plant, tree, opening door, backlighting, etc.

-- -- -- -- --

Request list I'm working on: Airi Sakura (COTE), Estelle (Majo), Le Malin (Listless Lapin), kabedon 3rd person pov

I used NaiBasil as a learning model to learn 19 illustrations of Maned wolf by Kemono Friends, includingofficial illustrations.

When using a Lora, if you set the minimum setting of 0.7, most of them come out well.

Below is a prompt for Maned wolf and basic clothes. It's better if you add Maned_wolf_(kemono_friends).

Basic clothes

• TATEGAMI,maned_wolf_(kemono_friends), wolf_girl, wolf tail,wolf ears,long_hair,multicolored_hair, orange_hair, (black_hair:1.2), (white_hair:1.4), animal_ears, hair_between_eyes, brown_eyes, gradient_hair, (short_twintails:1.3), extra_ears, large breasts, fang,
  
  crop_top,sleeveless_shirt, elbow_gloves, orange_bow, orange_bowtie,navel, (pantyhose:1.2), bare_shoulders, shoes, short_shorts,denim_shorts,fur collar, midriff, bowtie,arm_at_side,

trigger word : cybtruk

recomm. strength : 0.85 - 1.0 (works best),

if you like my stuff pls consider supporting me on Kofi

I used NaiBasil as a learning model to learn 32 illustrations of Koala by Kemono Friends, includingofficial illustrations.

When using a Lora, if you set the minimum setting of 0.7, most of them come out well.

Below is a prompt for Koala and clothes. It's better if you add Koala_(kemono_friends).

I trained basic clothes, nurse clothes. You can use the prompts below.

Basic clothes

• KOALACHAN,koala_(kemono_friends), animal_ears, twintails, brown_eyes, multicolored_hair, grey_hair,
  
  thighhighs, gloves,boots, elbow_gloves, apron, fur_collar,

Nurse costume

• KOALACHAN_nurse,koala_(kemono_friends), animal_ears, twintails, brown_eyes, multicolored_hair, grey_hair,
  
  nurse,white thighhighs,white gloves, hat,(pink shirt:1.3),(short sleeves:1.4),elbow_gloves, nurse_cap, clipboard, japari_symbol, stethoscope

Training model : ironcatfatetoon_v10

Sampling method : Euler a

suggest lora set 0.9-1.0 and add "Black Maria" in prompt

Have fun!

This is my second LORA

It is trained on beautiful mermaids pictures which i generated in midjourney

LORA reacts to words: white, underwater, blue, green,

You can find more words :)

Experiment LORA weights values from 0.5-1.2

Portrait images are generated very well

But that doesn't mean that you can only generate images of mermaids.

Сombine with my other LORAs and get interesting results

Create and enjoy!

For this one, I did differently than usual, and the results are much better than expected.

This is trained on a bunch of different variations of the female masturbation tag, with each variation being trained on the top 20 images from Gelbooru.

Results would probably be better if I did more than 20 images each, or more repetitions, but as it is, this took six and a half hours to train. So. Maybe I'll add to the data set and continue training over night.

Listed below:

• female masturbation, large breasts

• female masturbation, looking at phone

• female masturbation, masturbation through clothes

• female masturbation, mature female

• female masturbation, medium breasts

• female masturbation, multiple girls

• female masturbation, mutual masturbation

• female masturbation, on back

• female masturbation, on side

• female masturbation, on stomach

• female masturbation, orgasm

• female masturbation, panties around ankles

• female masturbation, panties around leg

• female masturbation, panties down

• female masturbation, pov

• female masturbation, public masturbation

• female masturbation, sitting

• female masturbation, small breasts

• female masturbation, smelling underwear

• female masturbation, solo

• female masturbation, squatting

• female masturbation, standing

• female masturbation, stealth masturbation

• female masturbation, watching

• female masturbation, yuri

• hand under clothes, hand under shirt

• hand under clothes, hand under skirt

• panties, panties aside

• panty pull, pulled by self

• breast grab, flat chest

• breast grab, large breasts

• breast grab, medium breasts

• breast grab, small breasts

• exhibitionism, clothes pull

• female masturbation, 2girls

• female masturbation, 3girls

• female masturbation, 4girls

• female masturbation, anal fingering

• female masturbation, caught

• female masturbation, clitoral stimulation

• female masturbation, fingering

• female masturbation, flat chest

• female masturbation, from above

• female masturbation, from below

• female masturbation, from side

• female masturbation, hand under clothes

• female masturbation, kneeling

hard to get a nice mankini with out error in all the check point so i made a Lora for it

enjoy it all :)

Lora of Jessica Rabbit. character from the movie "Who framed roger rabbit"

If you like my work you can buy me a Ko-fi, I would be grateful 🥰

What's the point of living if everything cannot be big chungus

--greatest philosopher

images were made using abyssorangemixs-2-Sfw

Olivier Mira Armstrong from the Japanese manga and anime Fullmetal Alchemist.

The trigger word is 'olivier'.

Weights of 0.5-1.0 are recommended.

If you want to wear something other than a military uniform, set 'Amestris military uniform' as a negative prompt.

If you want both eyes to show, set 'hair over one eye' to the negative prompt.

The thickness of the lips can be reproduced more consistently by adding 'Thick lips'.

To describe her character to those who don't know her, she is a 'major general in the national army', 'guards the country's northernmost border', 'first child of a prestigious military family', 'aged in her late 30s or older', 'single', 'always stern-faced and demanding' and 'subordinate-friendly'.

～～

日本の漫画・アニメ『鋼の錬金術師』のオリヴィエ・ミラ・アームストロングです。

トリガーワードは「olivier」です。

重さは0.5～1.0を推奨。

軍服以外を着せる場合は「Amestris military uniform」をネガティブプロンプトにしてください。

両目を出したい場合は、「hair over one eye」をネガティブプロンプトにしてください。

唇の分厚さは「Thick lips」を入れると、より安定して再現されます。

知らない人にキャラを説明すると、「国軍少将・国の最北の国境を守っている・名門軍人家の長子・30代後半以上・独身・常に険しい表情で厳しい・部下想い」。

Extracted from ToonYou

To get the full feel, take a look at the model

But combination with other models is also very interesting😃

The example image used MeinaMix v9

////////////////////////////////////////////////////////////////////////////////////

Do you like my work? check out my profile and see what else!

And A cup of coffee would be nice! 😉

////////////////////////////////////////////////////////////////////////////////////

Brief

This is the lora model for Sangonomiya Kokomi from Genshin Impact. Trained basing on anything V3, works on both anime and realistic checkpoint. It works well with additional network weight from 0.6~0.8. And yes, you can easily remove her costume or change to another as you wish.

For more info: https://huggingface.co/okingjo/Single-identifier_LORA_Model

How to use

You will need the following prompts to call out Kokomi:

sangonomiya kokomi, white thighhighs, white gloves, frilled sleeves, bow-shaped hair, vision \(genshin impact\), bow, white panties, detached collar, sandals

Note that some of the prompt can be removed accordingly

If you want to remove/change the costume, keep only sangonomiya kokomi, ( bow-shaped hair is optional), and put the other prompts from above in the negative if anything stays.

Please feel free to comment and let me know if there are anything you think could improve the training! I`m still learning and I`m open to feedback!

千景绘的画风lora，权重0.8-1

The painting style of Qianjing Painting, Lora, with a weight of 0.8-1

This artstyle is composed of bits of the helltaker style of artworks mixed with the OptionalTypo style. This adds a bit of depth to bodies, as well as intricacies to the color scheme of the helltaker style. This LORA is especially good at creating tomboy and nsfw artworks.

The trigger word for this LORA is johnny_numbnutz. I find a strength of 0.9-1.0 works the best for the intended output.

All pictures included are made using the hi-rez fix feature. Settings:
Denoising strength: 0.7,

Clip skip: 2,

Hires upscale: 2,

Hires steps: 50,

Hires upscaler: Latent

Lora model customization: click here

Animation series: Click here

Version Description

V100: Can change clothes, hair color, eye color, support various main models, support 512-1280 resolution

The model is made based on ch modal。

The image source and training code come from the Internet, and the model is only used for exchange of scientific research interests. If there is any infringement, please contact to delete, thank you.

图源和训练代码来源于网络，模型仅作为科研兴趣交流使用，如有侵权请联系删除，谢谢！

画像ソースとトレーニング コードはインターネットから取得され、モデルは科学研究の利益の交換にのみ使用されます.侵害がある場合は、連絡して削除してください。

Trigger word

Yune

Clothes

pink kimono, hair ornament\(origin\)

recommeneded Weight: 0.6-0.9

recommeneded model: Anything4.5

curren chan lora model based on new tv anime screenshot

Trigger words: 1girl, karen

Works well with anything model

出场于海虎终极系列的最后邪白,终极强者

这个模型设置有关键词，一定要填

tag:baiwunan,1boy,solo, male focus, manly,gloden belt,white ahoge，brown hair，thick eyebrows，solo, smile,white trousers，muscle lines，long ahoge，ahoge on forehead

Jessica Jones V1 - krysten ritter

All prompts are available on each pictures :-)

All my work with all prompt here (include NSFW)

patreon.com/Beerus_geek

Song dynasty style landscape model, Chinese aesthetics

A famous costume endorsed by SEKIRO.

This is a model that generates this SEKIRO COSTUME which trained by 50 img.

You can trigger by outfit sekiro,

Adjust the weight from 0.9~1.1. have fun （：3っ）∋

对于个人来说，发布这个LoRA是个很难的决定，因为涉及到的版权问题和潜在纠纷是在令人咂舌。

所以，我的这个作品只在C站供大家私人学习使用，任何违反法律的，商业的行为都是被禁止的。

C站独家的LoRA，如果出现任何版权纠纷，我都会立刻删除下架。

For myself, releasing this LoRA is a difficult decision, because the copyright issues and potential disputes involved are staggering.

Therefore, this work of mine is only available at Civitai for your Private Study and use. Any violation of the law or commercial behavior is prohibited.

The exclusive LoRA at here, if there is any copyright dispute, I will delete it immediately.

欢迎关注，发现更多好用的模组。

Welcome to Follow and discover more useful mods.

1. 炭治郎tanzhilang/Tanjiro Kamado/かまどたんじろう

2. 吾妻善逸wuqishanyi/Zenitsu Agatsuma/あがつま ぜんいつ

Please tell me if this model works properly. If it infringes, I will delete it immediately. Please forgive me.

It's finally here: the BLACKED LoRA. Please rate it if you - and your waifus - like it!

No trigger words, standard CFG, regular weight. Trained on NAI, should work with most models.

You don't need LoRA specific trigger words, just as long as you have include male tags in your prompt (e.g. 1boy, 2boys, multiple boys) and/or sex-related tags (e.g. sex, fellatio, gangbang). In some models you can increase the effect by using tags such as black man, interracial, etc. See the examples (non-cherry-picked generations).

The idea was to achieve good results with the least impact on style and composition as possible. You can use this LoRA in combination with other LoRAs (as long as those are not overtrained). Depending on the model, I suggest between 0.8-1.1 weight and your standard CFG.

Let me know if you encounter any issues, and provide feedback for improvement.

Post your own images, especially with different models and LoRAs!

This is a lora model of the character——史都华德/Steward。

权重一般用0.6比较好，想比较自由的0.5也不错，请务必配合使用controlnet，解锁更多姿势。

！请不要使用该模型进行商业用途

！Please do not use this model for commercial purposes

ps: 搭配画风Lora使用更佳！

Contacts：

QQ交流群

专注练男人！

Read: Due to censorship, the images created with the real LORA are hosted on reddit----->  https://www.reddit.com/user/PhotographRemote7402/comments/12qmmt4/smegma_lora/


The problem with this Lora is playing with the degree of deformation of the penis, since it is not created with "normal" penises it is normal for them to be deformed, but it is difficult for the deformation to come out as we wanted it and basically it generates more attempts by increasing the number of attempts that by selection, the intensity of Lora can be lowered a bit to seek to give more relevance to the composition of the image

smegma,<lora:smeg-13:1>---->Activacion
Top tags
penis (45)---> smegma
1girl (38)
breasts (34)
uncensored (33)
veins (32)
erection (31)
testicles (30)
large penis (30)
veiny penis (29)
hetero (29)
nude (27)
solo focus (26)
blush (25)
1boy (25)
nipples (24)
pubic hair (23)
long hair (22)
dark skin (22)
cum (20)
huge penis (19)
huge breasts (18)
male pubic hair (18)
large breasts (16)
makeup (15)
lipstick (15)
blue eyes (15)
short hair (15)
lips (15)
futanari (14)
solo (13)
precum (13)
looking at viewer (12)
dark nipples (12)
sweat (12)
censored (12)
tongue (12)
large testicles (11)
open mouth (11)
dark-skinned male (11)
tongue out (11)
oral (10)
fellatio (10)
interracial (10)
english text (10)
colored skin (9)
large areolae (9)
teeth (9)
smile (9)
thighs (9)
muscular (9)
black hair (9)
multiple boys (9)
navel (9)
dark areolae (8)
penis on face (8)
penis awe (8)
covered nipples (8)
mosaic censoring (8)
dark-skinned female (8)
red hair (7)
steam (7)
thick thighs (7)
brown hair (7)
cleavage (7)
heart (7)
multiple penises (7)
simple background (6)
pointy ears (6)
earrings (6)
blonde hair (6)
pov (6)
huge nipples (6)
gloves (6)
mature female (6)
group sex (6)

Trained on the animefull-final model for 11 epochs at a network dim of 32, resized to one of 16. I used a combination of official artwork by the game's creator Konjak along with fanart.

Hereby the tags I primarily trained on:

Character: robin, lips, ponytail/long hair

Outfit: red jacket/tank top, pants

Wrench: holding wrench, over shoulder, planted

You need a bit of gacha luck to make the ponytail work like in the original, its shape is pretty difficult for the AI to understand lol. Meanwhile her signature wrench doesn't work particularly well, which is to be expected with tools/weapons.

Using lips in your prompt will make her face look much closer to the official art, while keeping it out will make it more style neural.