Setsuna Yashahime

Things this Lycoris can do:

• Play you Violin at 4 am.

• Steal your dreams

• Give you spoilers to the two seasons of her show

• Multiple Outfits including Kimonos, Punkish outfits, her Naginata, and much more.

Trained using Holo's colab sets with dataset maker and Lycoris training.

This does NOT do NSFW as far as i'm aware, and I would BLOODY well hope not.

Idgaf that gelbooru has explicit images of her and her sisters, that's just wrong.

If you haven't watched Yashahime yet do it.

Also: we're not responsible for any wrong doings when you output with this lycoris, or any copyright infringements you make.

Do not RESELL this lycoris, but you're more than welcome to use it on generation websites as long as you're able to credit :D

This LoRA is not official or authorized.

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

LyCORIS of Brett Cooper

Dan Avidan aka Danny S3XB4NG

(yes in order not to trip the PG13 gods we fixed up the letters a bit, ti's a nickname not a literal act lol)

This model works on EVERYTHING because why? We wanted a lycoris that worked on illustration and realism models. So we did it.

He will likely tell you that he can't play on controller, amazingly tell you that Ninja Brian has an STD....

And probably sucking something dirty for 10 bucks.

This was trained using: https://colab.research.google.com/github/hollowstrawberry/kohya-colab/blob/main/Dataset_Maker.ipynb

And: https://colab.research.google.com/github/hollowstrawberry/kohya-colab/blob/main/Lora_Trainer.ipynb

Used 90% of Holo's default settings except for cranked the batch size to 4, cause I wasn't sure how long it would take to train it LOL.

Also for 90% of the images output: Dynamic Threshold, because we use Nocrypt's colab your images may be different.

If you like what we do come tell us here: https://discord.gg/Da7s8d3KJ7
Then find our Ko-Fi in our profile and dump a few bucks our way.

本模型为LyCORIS模型，请使用Show extra networks打开。

このモデルはLycorisモデルです，Show extra networksを使用してください。

This model is a LyCORIS model, please use Show extra networks to open it.

本模型内置两套服装，如果你想要星梨花的私服，请使用shifukuSRK tag，如果你想要「With My Heart」卡面上的服装，请使用ribbonSRK tag，如果服装不显示，请适当提高服装tag权重。

このモデルには2つの衣装が内蔵されています。星梨花の私服をご希望の場合はshifukuSRKタグを、「With My Heart 」カードの衣装をご希望の場合はribbonSRKタグを、衣装が表示されていない場合は衣装タグのウェイトを適切に増やしてください。

This model has two built in costumes. If you want the daily wear of Serika, please use the shifukuSRK tag, and if you want the costume on the 「With My Heart」 card, please use the ribbonSRK tag, and if the costume is not displayed, please increase the costume tag weight appropriately.

This LoRA is not official or authorized.

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

LyCORIS of Britney Spears

LyCORIS of Caity Lotz

LyCORIS of Carrie-Anne Moss

We will comply with the ANYCOLOR secondary creation guidelines.
https://www.anycolor.co.jp/guidelines/

It seems that no particular intention is indicated. I pledge to create and publish works that respect the spirit of ANYCOLOR content.

特に個別の意向が示されていないと思われます。ANYCOLORコンテンツの精神を尊重した作品を制作、公表していく事を誓います

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

Unfortunately it struggles with the Gauntlets, Clip Skip 1 seems to do it better but I honestly don't know! I'm considering to train a Negative TI for it like the bad hands one to improve on gauntlets but not sure yet. Negative Bad Hands TI also seems to help but not sure.

Trigger words for V1.20:

For original clothes: <lora:Viv120:0.8>(vi, original-clothes, gauntlets, pink hair, goggles on head, piercing)

For police uniform: <lora:Viv120:0.8>(vi, police uniform, policewoman, gauntlets, pink hair, hat)

For arcane version: <lora:Viv120:0.8>(vi, arcane, bandaged arm, red hair, hood, muscular female, piercing)

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

Sample images generated from Hassaku.

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

It was trained on 100 512x512 images scraped from the internet and 13 epochs using animefull_final.

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

We will comply with the ANYCOLOR secondary creation guidelines.
https://www.anycolor.co.jp/guidelines/

It seems that no particular intention is indicated. I pledge to create and publish works that respect the spirit of ANYCOLOR content.

特に個別の意向が示されていないと思われます。ANYCOLORコンテンツの精神を尊重した作品を制作、公表していく事を誓います

VACUUM BLOWJOB!!

This LoRA is not official or authorized.
https://vee-official.jp/guideline/
We will comply with the VEE Secondary creation guidelines.

This model was trained on Chilloutmix znd made with the LyCORIS method, so you'll need to install this extension made by KohakuBlueleaf to fully utilize it

you must install 2 extensions https://github.com/kohya-ss/sd-webui-additional-networks and https://github.com/KohakuBlueleaf/a1111-sd-webui-locon

The trigger token for this LORAs is: sksrachel
additional prompt if the trigger token is not enough: sksrachel, blonde hair, smile
example positif prompt:

RAW photo, portrait photo of "trigger word" as a (professor:1.3), (torn shirt:1.1), classroom, blackboard, 8k uhd, dslr, high quality, film grain, Fujifilm XT3

RAW photo, photo of "trigger word", as a soldier,wearing a (military uniform:1.2), cap, 8k uhd, dslr, high quality, film grain, Fujifilm XT3

portrait of "trigger word", as a (priest:1.3), smile, in a church, white robes, (holy cross locket:1.1), 8k uhd, dslr, high quality, film grain, Fujifilm XT3

Raw Photo "trigger word" with beard, helmet elder scrolls v nordic armor bethesda adamowicz illustration character design concept, unreal 5, daz, hyperrealistic, octane render, cosplay, rpg portrait, dynamic lighting, intricate detail, harvest fall vibrancy, cinematic volume inner glowing aura global illumination ray tracing hdr,

RAW photo of sksrachel, street smiling, backpack, ponytails, epic realistic, photo, faded, complex stuff around, intricate background, soaking wet, neutral colors, ((((hdr)))), ((((muted colors)))), intricate scene, artstation, intricate details, vignette

Best samplers are DPM++ SDE Karras, DPM++ 2M Karras, Euler A

Recommended weights for txt2img: 0.6-1 (best is 0.8 sometimes need 0.7)

Asuka from the Senran Kagura franchise

• Trained with anime (full-final pruned) model.

• 4 versions; 8 epochs, 8-1:1 (8 epochs with 1:1 weights), 10 epochs, 10-1:1 (10 epochs with 1:1 weights)

• Good results with 0.8+ weights.

• 2 activation tags: asuka_sk (for general info), asuka_kunoichi_uniform (for her Kunoichi uniform), and asuka_striped_bikini (for her standard striped swimsuit.) Try adding tags for clothing in the prompt for the last 2 activation tags just in case.

CC License

A LYCORIS Locon of Sophie Twilight from the Anime/Manga series, Tonari no Kyuuketsuki-san. You will have to install this extension made by KohakuBlueLeaf into your webui to fully use a Locon LORAs

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

This Locon was made with a collection of fanart images and anime screencaps from the show, she can be easily changed into multiple outfits and hairstyles like was shown off in the series whenever a part of her hair was turned to ash or her beautiful grey hair was played around like a doll's hair. She's very cute for a 360 year old otaku vampire

Compatibility

Trained on the NAI model, so it's pretty much compatible with ever NAI based Anime model mix that you can find on the internet

Recommended Weight

Always 1.0 by itself and 0.8-0.7 if you're mixing it with other LORAs.

Trigger Tokens

normsophie will be the token needed for her normal look, you may want to up the strength on some additional tags because even when I tried correcting the tags given to her training images and used multiple network ranks and learning rates there is some rng where her outfit will be a full on black one but as long as if you use

"very long hair, black dress, hair between eyes, white sailor collar, brooch, red ribbon, black pantyhose, tress ribbon, black tress ribbon, sidelocks, grey hair, red eyes"

you'll get her main look, not sure if it was the fanart look some people had with her outfit conflicting with the anime data that caused this but otherwise you'll get her main look 90% of the time

altsophie will be the token you use to allow for alternative outfits and hairstyles a lot more easier, if you notice one of her main outfit's clothing appearing in your outputs ensure you put one of the main tags from her main look into the negatives, didn't happen with my outputs but if it does there you go

Been awhile since I posted a LORA and my outputs on my other social media platforms so if you want to support me, like always just check my posted links here, I'll be thankful if you support me through all or one of my platforms and I'm looking forward to working on more LORAs o/

Trained on 29 images don't expect some good results.

trigger world - Nygglatho

strength - 0.6 - 0.9

Creepy Chan, is the online pseudonym of fashion model Allison Harvard, whose photos were shared on 4chan as early as 2005.

This is an alternative approach to my TI model (https://civitai.com/models/40608/creepy-chan-meme)

I used the process developed by @malcolmrey. Thanks so much for not only providing amazing models but also sharing the process how to do it!

Please let me know if you like it. That would be very much appreciated. Also let me know, what you think about Textual Inversion vs. LyCORIS.

LyCORIS of Cirilla Fiona Elen Riannon

LyCORIS of Demi Moore

LyCORIS of Donald Tusk

LyCORIS of E. Page

Will get retrained soon.

Some chocolate frog convinced me to make this LOCON model, as i try with all my loras they should work with any checkpoint

KARTRIDER：DRIFT ：KRIS（跑跑卡丁车：漂移 微笑）的lora-lycoris用模型 注意安装lycoris的插件。才可调用。

训练用的图 弄的较少 听话程度不算高，但是找到合适的词以后 效果还是很不错的。可以参考我上传的几张图。

需要加 chrios 触发词

怪盗kris的服装词语：fuzhuang

欢迎加入 跑跑卡丁车：漂移玩家交流群 qq：547483410

KARTRIDER: DRIFT: KRIS (KartRider: Drift Smiling) lora-lycoris model, please install the lycoris plugin before calling.

There are fewer pictures for training, and the obedience is not high, but after finding the right words, the effect is still very good. You can refer to several pictures I uploaded.

Need to add chrios trigger word.

Costume words for Phantom Thief Kris: fuzhuang

Welcome to join the KartRider: Drift player exchange group qq: 547483410.

ハチナイ本校のユニフォームLoHAです

学習素材が背景透過だったからか背景があまりよくない気がするので気が向いたらVerupする

商用利用できるクオリティではないと思うけど商用利用はやめてね

使用时除了触发词，你需要加入人物主体（比如1girl,2boys等）并对衣着进行描述（比如nude或者bottomless等）,此外你可以写一些tag来尝试增加稳定性、引导你想要的结果，这也同时取决于你的模型。下面我给出一个例子（这些并不是必要的，仅供参考，即使你不写其余tag也可以使用）：

In addition to trigger words, you need to add the main body of the character (such as 1girl, 2boys, etc.) and describe the clothes (such as nude or bottomless, etc.), and you can write some tags to try to increase stability and guide the results you want. It also depends on your model. Below I give an example (these are not necessary, just for reference, even if you don't write other tags can also be used):

double penetration, 1girl, 2boys, anal, vaginal, all fours, from side, bottomless, ass, penis, testicles, from side, mmf threesome, straddling

还有一个特殊视角可供使用，你可以加入“mating press, boy on top”来让上方的男性改变位置（我不知道如何描述，请看后面的例图）

There is also a special camera available, you can add "mating press, boy on top" to make the male above change position (I don't know how to describe it, please see the example picture below)

目前存在的问题：

1.下方的男性身体经常会消失一部分（但是他提供了一个阴茎这就足够了吧笑）

2.成功率有时不是特别高（尤其是第二种视角下）

3.有一点过拟合，请使用hiresfix纠正

currently existing problems:

1. The lower part of the male body often disappears (but he provides a penis which is enough lol)

2. Sometimes the success rate is not particularly high (especially in the second perspective)

3. There is a little overfitting, please use hiresfix to correct

darling in the franxx 我很喜欢的一个角色~

Loraの強度は0.5ぐらいにして使って下さい

【必要なプロンプト】

顔の再現は white hair,very long hair,fox ears, purple eyes, ponytail, hair ornament,parted bangs,

服の再現は japanese clothes,white kimono,

足元の再現は geta,tabi,

上記のプロンプトでだいたいの再現ができます

服関連のプロンプトを全て外せば着せ替えができます（重要）

gorstでNHK（幽霊みたいなマスコット）を召喚できます

学習モデルはAnyting V5 を使っています

The learning model uses Anyting V5

Please use Lora's intensity at about 0.5

[Required prompt]

Reproduction of the face is white hair, very long hair, fox ears, purple eyes, ponytail, hair ornaments,

Reproduction of clothes is japanese clothes,

The reproduction of the feet is geta, tabi, white kimono,

You can roughly reproduce the above prompt

If you remove all clothes-related prompts, you can change clothes (important)

Please see the sample image for other detailed parts.

自训练微调大模型提取locon

LyCORIS of Jennifer Lawrence

LyCORIS of Katarzyna Warnke

LyCORIS of Maitland Ward

LyCORIS of Maja Staśko

LyCORIS of Mary Lynn Rajskub

LyCORIS of Megyn Kelly

LyCORIS of Michelle Rodriguez

LyCORIS of Miley Cyrus

LyCORIS of Millie Bobby Brown

Hey hey! Since this is my first upload for today I'm gonna share an update: I'm still doing the guide, the audio is all done, I'm mixing the video and preparing additional text materials. I think my deadline of tomorrow (Wednesday) should be fine :)

The tutorial will be included along with Morgan LyCORIS

自分自身のイラストをLoRA化しました。

使いたい人はどうぞご自由に。

※このファイルを使用した場合のいかなる損害も私MistRalは負いません。

I LoRA-ized my own illustrations.

If you want to use it, please feel free.

/I, MistRal, will not be responsible for any damage caused by the use of this file./