權重:0.6-0.8搭配Cutoff效果更好

Weight 0.6-0.8,Using Cutoff works better

Requires extension:

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

Recommended tags are:

torako, 1girl, white hair, white kimino, long hair, blue flower, blue eyes, bare shoulders, pattern, white pantyhose

Rosalina LoHA | 5 Outfits in 1

This is a LoHA Lycoris and you need this Extension for this to work. My first LoHA Model and I'm happy with the results. Credits go to @EDG for helping me out. Thank you.

Weight can go from 0.7 up to 1.0. I recommend 0.9. ControlNet will streamline full body shots much easier, or more complex poses. Works with almost every Style, every alternative Outfit, Photorealistic Models and NSFW LoRA's.

The Main triggerword is Rosalina. You can add a second trigger tag like Rosalina_Biker, Rosalina_Casual, Rosalina_OG, Rosalina_StarryNightDress and Rosalina_Striker. You can also only use second trigger word, to give other Char Rosalina's Outfit (see example). If you wanna go for other Outfits for Rosalina, then use Rosalina_Casual.

This will be updated soon with other Outfits from Rosalina. I saw Sommer Outfit and some other Outfits in Mario Kart Tour. This will be included in the next Update.

Trigger Word:

Rosalina

Outfit 1: Biker

Rosalina, Rosalina_Biker

Outfit 2: OG

Rosalina, Rosalina_OG

Outfit 3: StarryNightDress

Rosalina, Rosalina_StarryNightDress

Outfit 4: Striker

Rosalina, Rosalina_Striker

Outfit 5: Other's

Rosalina, Rosalina_Casual

Bonus: Outfit on other Character (Note: Do NOT use Rosalina and you need to adjust weights/prompts)

Rosalina_Striker / Rosalina_StarryNightDress / Rosalina_Biker / Rosalina_OG, (other Char prompt/LoRA)

How to use LoHA's in auto1111:

• Download LoHA

• Copy the file to stable-diffusion-webui/models/lora

• Select your LoHA and use the triggerword

• Make sure to change the weight if needed (by default it's :1)

• Have Fun!

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

Butterfly Lingerie

Beautiful butterflies...

This LORA is Hiiragi Kotoha from Hachigatsu no Cinderella Nine.

Kotoha is a character in the smartphone game Cinderella Nine in August.

The player becomes the manager of the girls' baseball team and gives them instructions.

She is a good defensive left fielder and a good cook.

inital tags:

"hiiragi kotoha, long hair"

main costume:

"baseball uniform, pink neckerchief, black undershirt, fingerless gloves, belt, pleated skirt, bike shorts, bike shorts under skirt, black socks, sneakers"

summer school uniform:

"serafuku, blue necktie, plaid skirt, black socks, loafers"

winter school uniform:

"blazer, school uniform, neck ribbon, plaid skirt, black socks, loafers"

colabo costume:

"side ponytail, white shirt, white skirt, thigh strap, blue under shirt, blue ribbon, wristband, star hair ornament, black socks, sneakers"

kimono costume:

"hair flower, white kimono, fox tail, fox ears, fox girl"

LyCORIS of Sadie Sink

This LORA is Takanashi Yuzu from Hachigatsu no Cinderella Nine.

Yuzu is a character in the smartphone game Cinderella Nine in August.

The player becomes the manager of the girls' baseball team and gives them instructions.

Her position is shortstop, and she is the player's junior and childhood friend.

inital tags:

"takanashi yuzu, ponytail, hair bow, green bow"

main costume:

"baseball uniform, pink neckerchief, black undershirt, fingerless gloves, belt, pleated skirt, bike shorts, bike shorts under skirt, black socks, sneakers"

summer school uniform:

"serafuku, yellow necktie, black socks, plaid skirt, black socks, loafers"

winter school uniform:

"blazer, school uniform, neck ribbon, plaid skirt, black socks, loafers"

colabo costume:

"hair bow, white bow, midriff, tied shirt, orange skirt, black gloves, black thighhighs, sneakers"

kimono costume:

"yukata"

LyCORIS of Sarah Hyland

LyCORIS of Sasha Grey

LyCORIS of Trieste Kelly Dunn

LyCORIS of Uma Thurman

LyCORIS of Willa Holland

LyCORIS of Zendaya

Created from 34 images from various victorias secret catwalk model shoots

A request from net friend.

Role of HonkaiImpact3rd, and it's a ninja costume.

Weight 0.8~1.0 ,and you can ADJUST prompt to reductive character

TRIGGER:

1.BASE: sakura neon;

2.OG Costume: armor,mask on head

PS:too more 3D img...

Khoudia Diop, also known as the Melanin Goddess, is a Senegalese fashion model and actress.

This is the LyCORIS version trained on the dataset I also used for the TI version (https://civitai.com/models/38077/khoudia-diop-model)

Please let me know which you prefer!

a style

Recommended parameters:

推奨パラメータ：

推荐参数：

Model: aniflatmixAnimeFlatColorStyle_v10

Vae: 840000

Clip skip: 2

weight:0.7~1

Sampler: DPM++ SDE Karras | DPM++ 2M Karras

Steps: 20

Size: 512x512|512x768|768x512

CFG Scale: 7

☆Heads up: we’re using the v1.0 base model, not the latest v2.0, 'cause the v2.0 style is kinda retro and doesn’t vibe with what we’re going for.

☆注意：ベースモデルはv1.0を使っています。最新のv2.0ではなく、v2.0の画風はレトロ調になってしまい、期待した効果と合わないからです。

☆注意：底模用的是v1.0，而不是最新的v2.0，因为2.0版的画风会偏复古，与预期效果不太一致。

————

Instructions:

Only supports flat-color base models, and using other base models may cause oversaturation. You may need to lower the weight or use a layer weight plugin.

Good at drawing scenery (mountain, tree, cloud, sun, moon, star), good at wide image.

Fan art, please do not use for commercial purposes.

説明：

フラットカラーのベースモデルのみ対応しており、他のベースモデルを使うと過飽和になる可能性があります。ウェイトを下げるか、レイヤーウェイトプラグインを使う必要があるかもしれません。

風景（山、木、雲、太陽、月、星）や横長の画像が得意です。

同人作品ですので、商用利用はご遠慮ください。

说明：

仅支持平涂类底模，且使用其他底模有可能会过饱和，需要拉低权重或使用分层权重插件。

擅长画景（山、树、云、太阳、月亮、星），擅长横版。

同人作品，请勿商用。

————

If you have any ideas, suggestions, or questions during use, you can ask them on Discord ↓

ご利用中に何かアイデアや提案、質問がございましたら、Discordにてお問い合わせいただけますようお願いいたします↓

https://discord.gg/tVpJMhjUwx

使用中有任何想法&建议&问题，可以点上面的链接进Discord里提问，也可以在哔哩哔哩提问↓

https://space.bilibili.com/1328667069

Also, we made a tutorial called ‘AI Painting Tutorial-How to Create Stunning Images with Stable Diffusion’, which can help some creators who need to level up their image-making skills. You can find the link in the FAQ section on Discord.

また、『AI Painting Tutorial-How to Create Stunning Images with Stable Diffusion』というチュートリアルを作りました。画像作成のスキルを上げたいクリエイターの方に役立つと思います。リンクはDiscordのFAQにあります。

此外，我们制作了一个教程“AI绘画教程-喂饭级提高出图质量全流程（Stable Diffusion）”，可以帮到一些需要提高跑图质量的创造者，链接在Discord的FAQ里就可以看到。或者直接点击下面的链接。（也有视频版，但也许文字版更清晰一点，可以先使用ctrl+f搜索关键字找找有没有你遇到的问题，希望可以帮到你~）

https://www.bilibili.com/read/cv23237755

————

If you’re new to LoHa, please read this:

This model was trained with LoHa, so you need to install the WebUI extension. Trust me, it’s worth it.😉

https://github.com/KohakuBlueleaf/LyCORIS

LoHa model path: models/Lora

Usage: Click on the 🎴 button under “Generate” in the WebUI, choose the Lora tab, select the model, and adjust the weights. It’s as easy as pie.🍰

LoHaを初めて使う場合は、こちらをご覧ください：

このモデルトレーニングにはLoHaが使用されており、WebUIの拡張機能をインストールする必要があります。

https://github.com/KohakuBlueleaf/LyCORIS

LoHaモデルのパス：models/Lora

使い方：WebUIの「生成」の下の🎴をクリックし、Loraタブを選択して、モデルを選択して、重みを変更してください。

第一次使用LoHa请看：

本模型训练使用了LoHa，需要安装WebUI的扩展。

https://github.com/KohakuBlueleaf/LyCORIS

LoHa模型路径: models/Lora

使用：点击WebUI“生成”下面的🎴，选择Lora 选项卡，选择模型，更改权重。

————

Buy us a cup of coffee and we’ll spill the beans.

https://ko-fi.com/co_subcreation

Or Fanbox

https://co-subcreation.fanbox.cc/

Or爱发电

https://afdian.net/a/Co_Sub_creation

Your support is the fuel that powers our world-building. Your ideas are the bricks that shape our world-making!

あなたの応援で、世界は生まれ変わる！あなたのアイデアが、世界を作り上げる原動力になるよ！

您的支持将成为创造世界的动力，您的想法将成为构建世界的基石！

Trigger word

• m2

• headgear

When reproducing a costume, m2 susually sufficient. However, adding the following tags may potentially lead to better results.

long hair, black_hood, black_cape, blue_jacket, long sleeve, white_boots, black_sox, white_shirt, attache_case, shoulder_bag, bag, medical_bag, 

This LyCORIS has been trained after carefully selecting 37 images. All photos have been enhanced to make the most of them. I hope you like it! Any feedback would be highly appreciated!

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

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

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

对文字的还原不是很好

The text restoration is not good

一样没有触发词，只需要写人物就可以了

no trigger words, just set the characters

分享一个可以达成双人物lora的简单方法，如果你有更好的欢迎分享在下面

Here is a simple way to achieve double-character lora. If you have a better way, please share it below

1. 单独调用一个人物lora生成图片，让一个人物符合你的期望

   Use a single character lora to generate a picture

   

2. 发送到图生图局部重绘，蒙版选中另一个人物，然后删除前一个人物lora，换成另一个人物lora
   Send it to i2i, select another character as a mask, then delete the previous character lora and replace it with another character lora

   

3. 生成图片完成角色替换
   Done

   

LoCon trained on 133 artworks by Himaruya Hidekaz, author of Hetalia

I'd keep it under 1.0, gets pretty good results at .5

Made to learn loRA training, this is my first loRA so may or may not be ass

Made a loRA before this but this one's just all around better

I don't think it has any triggers? I didn't use any at the very least

”当时只是做着玩的，觉得有意思就发出来了，没想到没火“

”It was just for fun at the time, and I posted it when I thought it was interesting, but I didn’t expect it to be unpopular“

没有触发词，给出人物数量然后随机生成即可

There is no trigger word, just give the number of characters and generate them randomly

你可以加上比如naked sheet之类的词来让它稳定一些

You can add words like naked sheet to make it more stable

如果你一定想去除画风的过拟可以尝试分层

If you really want to get rid of the oversimulation of the painting style, you can try layering

1,0,0,0,0,0,0.2,1,1,1,0,0,0,0,0,0,0

Note #1: As this is a Lycoris/Locon file, you will need https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris extension to use the <Lyco> tag. If you want to use the prompts like I do, specifically being able to use <Lora> instead of <Lyco> (only effects ease of use), you will need this as well https://github.com/KohakuBlueleaf/a1111-sd-webui-locon. Either of these work, you will just need to choose between using <Lyco> and <Lora> prompts.

What does it do you ask? Uh oh, your waifu has been captured and defeated in her fight against the villains! Now they've captured her and she's... Well, she's not going to have a great time. Or maybe she will? It really just depends on how you prompt it. Overall, the more weight you add to this Locon, the more broken and defeated your waifu will be in the image. A light weight will just guide it. A heavy weight will let it take over. And then you can prompt it however you want. Importantly, I did decide to train it where usually there is another character in the scene, so solo images can be done but require more effort. If people are interested in me emphasizing more solo usage let me know.

How to use it:

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

0.5 to 1 depending on model/lora/etc combination. And how heavy you want this Locon to take the reigns.

Main Triggers

BrokenM - Trigger word to focus the concept.

After that... Whatever you want!

hetero, penis, 2boys, multiple_boys, bruise, chain, injury, rape, cum, facial, ryona, defeat, uncensored, after rape, blood, restrained, bite_mark, threesome, group sex, mmf_threesome, strangling, asphyxiation, grabbing another's hair, tongue out, tongue, ahegao, straddling, tears, pee, peeing, interracial, defloration, piercing, leash, bdsm, clitoris piercing, bound, pussy piercing, bondage, arm grab, girl on top, sex, breast grab, torn clothes, cowgirl position, crying, finger in another's mouth, syringe, drugs, after sex, breast press, vaginal, spread legs, cum in pussy, duel monster, large beasts, monster, multiple boys, public indecency, crowd, auction, ryona, bound, restrained, collar, leash, piercing, dildo, sex toy, object insertion, pussy juice - Really... It's up to you and what you want it to do. There are definitely a ton of other common prompts you can use to help define the scene

Extra, Extra!: Should be well trained and usable in most models. Images shown for example.

Picture Descriptions:

1: AbyssOrange using just this Lora.

2: AnythingV3 combo'd with my released Anime Officer Locon.

3: AnythingV3 combo'd with my released Gothic Locon.

4: AnythingV3 using just this Lora.

5: AnythingV3 combo'd with my next released Evil Mask Locon.

6: AnythingV3 combo'd with my in progress Ann Takamaki (P5) Lora.

7: AnythingV3 combo'd with my released March 7th (Honkai) Lora.

8: AnythingV3 combo'd with my released Makoto (P5) Lora.

9: AnythingV3 using just this Lora but very lightly.

10: RevAnimated using just this Lora.

How was it trained?

It was trained on animefull_final with 32 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?

Unlike my previous ones, I've had to lower the total amount of images in my dataset from what I initially picked, as I decided to go the high quality route. I think I'll want a v2 that extends potential usage.

What's next that isn't this?

More character Loras. (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, Lucoa from Dragon Maid, Laharl-chan from Disgaea, 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, Bound Yuri (two girls tied up, breast to breast and likely kissing), Sultry Succubus Transformation, Tied up with vibrators (basically the waifu tied up with vibrators and being overwhelmed with pleasure), Beta Censorship (Automatically censors any images you create to make them BetaSafe), being stepped on, Robotification, 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), 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, 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 a Delinquent (The delinquent girls in japanese schools specifically), GoodCop/BadCop (Locon with outfits for being a good and for being a bad cop), Maid (both normal and naughty variants), Biker Chick (ride a motorcycle and look like you should), Race Queen (Chiling sexily on a car), forest fairy, Anime Officer, barbarian.

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

Second Lazy Style Lora. This time much bigger. It's still focused on anime-like style, so, difference from GPTS1 is not very big.

It is made on the base of extra lazy tagging way, which involves only barebones editing of auto-tagged images. This dataset consists of 404 images.100% SFW.
Additional experimental tag filtering was applied(maybe i'll release that project in future).

Of course, main theme of dataset is anime waifus, but considering dataset size, there are bunch of other themes to ease normalization of style to other themes, which led to better cooperation with tags that were not trained.

Generally, this is not supposed to be used for anythin complex, because it was not tagged in a way to be, but you still can try and show me your results :3

Same as GPTS1 Lite - it will have a lot of bleed, but i noticed that it's less in this one.

If you want it to be editable and still have some style effect, try weight from 0.5 to 0.75. That should make it editable and still be looking good :D

Generally, this dataset processing used my custom tag filtering experiment, testing of which on GPTS1 led to increased editability, so, it might not be as hard to edit this one, but still, main purpose is just to get that bit of "wow" effect style.

Was not able to make it work with dynamic normalization, so, only rank 64(that did work for some reason, even though it's supposed to die with Adan d-adapt, as my experience with other trainings suggest).

Garie Tuman is a character of Senki Zesshou Symphogear GX.

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

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

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

Also, EasyNegative is used.

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

• blue dress, black dress, puffy sleeves, long sleeves, bloomers, striped bow,mary janes

Updates V1 → V2

I used more traning images and improved the learning method.

• Size : 114MB　→　28MB

• We can undress or change the costume more easily.

• It's compatible with more models.

Enjoy!

【日本語版】

ガリィ・トゥーマーンは戦姫絶唱シンフォギアGXに登場するキャラクターです。

おすすめweight：0.7～1 (0.7が一番いいかも)

アップした全ての画像はanything-v4.5で作成しました。

使用したVAEはkl-f8-anime2.ckptです。

また、EasyNegativeも使われています。

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

• blue dress, black dress, puffy sleeves, long sleeves, bloomers, striped bow,mary janes

アップデート V1 → V2

教師画像を増やし、学習方法を見直しました。

• サイズ : 114MB　→　28MB

• より簡単に脱がせたり服を変えたりできるようになりました。

• 相性の良いモデルが増えました。

楽しんで！

This LyCORIS has been trained after carefully selecting 41 images. All photos have been enhanced to make the most of them. I hope you like it! Any feedback would be highly appreciated!

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

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

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

I plan to add an additional non-LoHA version as current v1 focuses more on coherency, requiring more prompting. Some might prefer a simple strong style so I will support it.

Initial Version:

A LoRA for captain_kirb's Style based on Any_Lora.

Shouldn't need any trigger words, however adding 'captain_kirb' or 'captain_kirb style', may force the style more. Works best at weight 0.8-1.0. Clip Skip 2, however it is not required to generate good images. Trained on booru tags. Mainly NSFW, can still do SFW.

Likes to generate original/non-existant characters with simple prompts.

For best results, or results similar to mine, most anime-style models should work. Additionally, it is most recommended to use badv5/EasyNegative textual inversions.

Should work fine with other LoRAs.

Examples done using many different models. Follow prompts to attempt replication and use given negatives for better generations.

Usage:

• CFG Scale: 7-9

• LoRA Weight: 0.8-1.0

• Should work fine for less detailed prompts. High-res fix is highly recommended

• Negative I use: bad, incorrect, missing, covered, inserted, vaginal, poor, low, normal, natural, complex, less, small, incomplete, solo, mutated, extra, multiple, extreme, flat, wrong, lost, limited, careless

This model is to recreate the character "Holly" From Monster Rancher

Widowsuit LyCORIS

This is a model to recreate the widowmaker costume on any character.

The main tag is: widowsuit, arm tatto, tattoo

Other tags are: back tatto, head-mounted display (this one works sort of)

I recommend a Weight of 0.9 for widowsuit and if you use another character lora no more than 0.7

You can type (pink bodysuit),(red bodysuit),(purple bodysuit),(yellow bodysuit), (blue bodysuit) or another color to change the color of the suit, but it doesn't always work

It also contains the tag: white eyes, no pupils

Clarification:

Tends to do a lot of "salute"

Link to use in PixAi: https://pixai.art/model/1611227478668030874

Landy from Epic Seven, trained as a LyCORIS network.

Requires extension for use:

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

Recommended tags are:

dslandy, pink hair, long hair, braid, yellow eyes, android, barcode tattoo, military uniform, white dress, cape, hat, elbow gloves, red thighhighs

Check out my HuggingFace repository:

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

Check out my Pixiv:

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

Yuna FFX Cosplay Lycoris (TEST Version)

I'm experimenting with LoCon parameters for attire and concept, depending on your model / prompt might hit or miss a lot.

Works best with FantasticMix model. ( v2.0 is better imo )

fantasticmix_real - v3.0 | Stable Diffusion Checkpoint | Civitai

Trigger words : yffx , yuna (ff10), miko outfit, hair ornament, sash, hakama skirt, blue skirt, hakama, detached sleeve

Recommended Weight : 0.6-0.8

For best result use Dynamic CFG Threshold:

• Mimic = 10 | Clamp Percentile 99.5

• Min. Mimic Scale = 0, Cosine Up

• CFG Scale Sch. = 15, Half Cosine Down

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

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

This LyCORIS has been trained after carefully selecting 45 images. All photos have been enhanced to make the most of them. I hope you like it! Any feedback would be highly appreciated!

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

CFG scale should be between 4-7.

LyCORIS weights should be between 1-1.4

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

Trish DMC5 Cosplay Lycoris (TEST Version)

I'm experimenting with LoCon parameters for attire and concept, depending on your model / prompt might hit or miss a lot.

Works best with FantasticMix model. ( v2.0 is better imo )

fantasticmix_real - v3.0 | Stable Diffusion Checkpoint | Civitai

Trigger words : trish (devil may cry), tdmc, blonde, long hair, cleavage, bustier, pants, choker

Recommended Weight : 0.6-0.8

For best result use Dynamic CFG Threshold:

• Mimic = 10 | Clamp Percentile 99.5

• Min. Mimic Scale = 0, Cosine Up

• CFG Scale Sch. = 15, Half Cosine Down

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

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

Re-training this character from Azur Lane sharing my birthdate fared better than Nautilus since her dataset had a few more images than her, therefore her 'Fluttering Fanfare' costume came out almost perfectly and backgrounds will look fine. Sometimes it's better to put 'red shirt' in negative prompt since it might generate her in a red kimono/yukata(?) and certain artstyle LoRAs I tested with her tend to generate her without her cat ears and tail for some reason, so I guess it's best to keep her animal tags all the time too I guess.

LoCon copying the style of the banners from Honkai: Star Rail.

RECOMMENDATIONS:

Generate a 768x768 image and then send it to img2img and upscale it. When upscaling, you may need to remove LoCon or reduce LoCon weight to get the best results.
This LoCon does not require a trigger word.

Good luck!

Engorged Veiny Breasts V1.1b/c - fC (lora/lycoris)

by fitCorder

V1.1b - Stanard Lora version

https://civitai.com/models/57391?modelVersionId=63517

Lycoris / Locon (V1.1c) model requires Lycoris support addon a1111.

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

No faces in the training data. Model tends towards engorged closeups as a result. Prompting out-of-frame as a negative prompt tends to help.

Keywords

veiny breasts, engorged-fc, breast veins, veins  <lora:fC-engorged-v1.1b:1>