The generation of hands and steps is very problematic, it is recommended to use'hands outside the frame, feet outside the frame'.

I will share the Lora file I made here as soon as possible: https://t.me/+Q1p3vUDvVPQ2ZDRl

Trained on 165 pics.

*Updated with 75 more bunny pics.

Trigger word: toki

Battle suit : (sleeveless turtleneck leotard:1.1) ,elbow gloves, headset, boots,maid headdress,

Maid : gloves, bow, dress, maid apron, long sleeves ,frills ,bowtie ,black dress

Bunny: playboy bunny ,highleg leotard ,rabbit ears, wrist cuffs,(very_long_hair:1.1), multicolored hair, rabbit tail,

All pictures are generated on anything4.5 with some Artstyle LoRA & Inpaint fixed details.

Supports all traits, you can check those at Opensea or other marketplaces

LoHa of Mejiro Ramonu (umamusume). Please install extension on WebUI.

After installing extension, put LoHa file in your sd/models/lora folder and write LoRA notation in prompt to applying LoHa. <lora:loha_name:weight>

All uploaded images are generated by AOM3A1

Recommended options

• LoRA weight 0.7 ~ 0.8

• Trigger words

  • Base

    • mejiro ramonu \(umamusume\)

  • Racing competition costume :

    • ear piercing, white glove, single glove, wristband, white shirt, blue ascot, open clothes, white jacket, cropped jacket, long sleeves, high-waist skirt, black skirt, frills, single thighhigh, black thighhighs, asymmetrical legwear, white footwear, black footwear, high heels, asymmetrical footwear, mismatched footwear

  • Tracen summer uniform :

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

  • Tracen winter uniform :

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

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

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

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

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

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 64 images consist of

  • 4x low quality 20 images

  • 8x medium quality 15 images

  • 16x high quality 29 images

  • => 1 epoch = 664 images

• replaced character feature tags with mejiro ramonu \(umamusume\)

  • 1girl, multicolored hair, black hair, solo, yellow eyes, grey hair, mole under eye, alternate costume, parted bangs, single hair bun, forehead, red eyes, braid, black eyes, animal ears, simple background, horse tail, braided bun, grey eyes, purple hair, bangs, blue eyes, horse girl, purple eyes, two-tone hair, hair bun, medium hair, pink eyes, brown eyes, streaked hair, horse ears, brown hair, white hair

  • I think these tags are the essentials of Mejiro Ramonu or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

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

And, special thanks to my best friend. DH.

A ufotable screencap, will get retrained as a LoHA soon enough.

Recommended weight for the LoCon: 0.3~0.45 if you didn't fate artstyle to take over the character you want, keep it in your mind that most of Dataset was gathered from fate, i recommend 0.3 the most. It works good, for upscaling use ultrasharp-4x with denoising 0.3 & Tiling [Multidifussion] Method

0.5 and above if you want to do a chara in fate style

weight: 0.6-

prompt:

• driving - sit in driver's seat

• riding - sit in back seat

If you create any good images using this LyCORIS, please consider sharing them by leaving a review. Thank you!

• Pregnant, Multiple_girls to trigger, or you can use 3girls / 4girls / 5girls / 6+girls , please note that AI is not good at counting.

• This LyCORIS may ruin the art style, so it's better to lower the weight and use img2img to upscale.

• You can also generate Multiple_girls without Pregnant bellies, but since all the training images are pregnant girls, the results may not be very ideal.

• Known issues: Sometimes there may be extra belly buttons or belly buttons mistaken for nipples, or extra hands covering the stomach. In some checkpoints, if the weight is set too high, it may make the characters look too young.

• I used ClearVAE to generate sample images.

Surf bodybuilder Takuya's LoHa.

This LORA is Hachisuka Goemon from Oda Nobuna no Yabou.

She is a ninja who was a companion of Kinoshita Toukichirou.

Usually with she aiding Yoshiharu the protagonist in all his missions.

inital tags:

"hachisuka goemon, short hair"

main costume:

"hat, scarf, black scarf, japanese clothes, elbow gloves, fingerless gloves, gauntlets, fishnet pantyhose, boots"

The hat is very strongly rooted in her.

You may want to use Anything or 7thLayerAbyss or AOM3A.

Mitsuki Takahashi from the ane to osananajimi wa naka ga ii! ~ ninben tsuitenaishi eroge.

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

• 6 versions; 5 epochs, 6 epochs, 7 epochs, 8 epochs, 9 epochs, and 10 epochs.

• Obtain good results with ALL, MIDD, OUTD, and OUTALL LoRA weight blocks, with 0.5-0.8 weights

• Tip: Use the ALL0.5 LoRa weight block to make pre-made images ready to be enhanced throught img2img.

This LoCon is meant to improve guns, it is not the definite cure to bad guns, but it will help to make better guns.

Guns that were in the dataset are: handguns, submachine guns, assault rifles, rifles, sniper rifles, shotguns, light machine guns.

No unique trigger word for them other than their individual tags.

All pics were generated using Aurora.

LyCORIS of Courtney Ford

LyCORIS of Kiernan Shipka

LyCORIS of Lara Pulver

LyCORIS of Mena Suvari

LyCORIS of Saoirse Ronan

LyCORIS of Shailene Woodley

LyCORIS of Tamsin Greig

LyCORIS of Tessa Thompson

LyCORIS of Yvonne Strahovski

LyCORIS of Zoë Kravitz

LyCORIS of Zoe Saldana

What does it do you ask? Time to strap on your waifu to brainwashing and let it slowly reprogram her in whatever ways you want. Obviously, it would only be for moral and healthy things, like enabling her to stop smoking or to enjoy the taste of vegetables. We all want to protect them and nothing else. If people like this but want even more flexibility, I might try making a v2 that delivers more variations.

How to use it:

(I use Sampling Method DPM++ SDE Karras and Upscaler R-ESRGAN 4x+ Anime6B). 1.0 for the strap-em-in to a brainwashing machine. 0.6 to 0.8 if you want more flexibility in positioning and prompt flexibility.

Main Triggers

BrnWshM- The main trigger word and should be used most of the time.

Highly Recommended Tags

head-mounted display OR helmet - To make sure the programing goes nice and safely of course...

Some Useful Tags for a Machine:

sex machine, vaginal object insertion, object insertion, bondage, stationary restraints

If you want sex as part of itL

1boy, 1girl, hetero, vaginal, rape, sex, straddling, penis

Other Fun Tags To Try!

dark persona, evil grin, standing, mind control, corruption, pussy juice, tongue out, bdsm, tattoo, mask, saliva, blush, restrained, lactation, forced orgasm, sex toy, cum, pussy juice, barcode, barcode tattoo, tattoo, pubic tattoo, sweat, tentacles, piercing, cable, glowing, electricity, trembling, milking machine, heavy breathing, pubic tattoo, tube, h eadset

And whatever else you want! (I personally like glowing and electricity)

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

Picture Descriptions:

1-4: AnythingV3 with just this Lora and some light prompting for variability.

5: AnythingV3 combo'd with my in progress Viper (Nikke) Lora.

6: AnythingV3 combo'd with my released Tohru (Dragon's Maid) Lora

7: AnythingV3 combo'd with my released Rozalin (Disgaea) Lora.

8: AnythingV3 combo'd with my in progress Racer Queen Locon.

9: AnythingV3 combo'd with my in progress Mind Flayer Mindwipe Locon. (I found this one funny as it helped transform my mind flayer into a mech version >.<)

10: AnythingV3 combo'd with my in progress Blacked Locon.

11: AbyssOrange using just this Lora.

12: RevAnimated using just this Lora.

13: PerfectWorld 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?

I kind of want to make a v2 with more variability but that would obviously take time from other work. LMK on opinions.

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, 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), 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), 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, 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 a Goth Chick, 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 here.

First time making a LoKr very limited dataset.

(notice: Civitai changes the name of the lokr/lora if u copy prompt from the images it won't be exactly the same)

The model is trained with clip skip 1 on top of ACertainty

The locon is extracted from the full model

The original one seems to give worse result when applied to certain base model so I updated with one with smaller compression ratio

The associate hugging face repository contains other versions of models https://huggingface.co/alea31415/watayuri

Trigger words:

• ChibanaSumika

• ChibanaSumika, glasses

• KoshibaMai

• MamiyaKanoko, glasses

• MamiyaKanoko, hair bun

• NishideraNene

• ShirakiHime

• ShirakiHime child

• YanoMitsuki

• YanoMitsuki child

There are very few images of Nene and Sumika without glasses for now so we probably cannot get any reasonably good results.

Moreover, the model is only trained with screenshots for the moment, so all the images are tagged with aniscreen. You can put it in either positive or negative (putting it in negative could be better when switching to some base models).

Characters are separated by ;.

You can get the clothes correctly with general prompts like green school uniform or brown coat.

--

I don't need support or credit, but I would be glad to know that you are using the models I trained and find it useful.

Moreover, I would like to advocate for more franchise models.

You can take a look at my workflow https://github.com/cyber-meow/anime_screenshot_pipeline if you are interested.

I just want to spread the fact that there is no reason to encode a single concept in each lora.

LoHa of Narita Top Road (umamusume). Please install extension on WebUI.

After installing extension, put LoHa file in your sd/models/lora folder and write LoRA notation in prompt to applying LoHa. <lora:loha_name:weight>

All uploaded images are generated by AOM3A1

Recommended options

• LoRA weight 0.7 ~ 0.8

• Trigger words

  • Base

    • narita top road \(umamusume\)

  • Racing competition costume :

    • ear cover, star hair ornament, brooch, sleeveless dress, purple dress, white bow, mismatched gloves, white gloves, black gloves, shoulder cutout, puffy long sleeves, detached sleeves, see-through sleeves, frills, shorts under dress, short shorts, purple shorts, thigh strap, knee boots, purple footwear

  • Tracen summer uniform :

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

  • Tracen winter uniform :

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

Use a model derived from or mixed with animefull model.

• Settings

  • Stable Diffusion WebUI by Automatic1111

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

  • about 20 steps, CFG scale 3.5~6.5

  • I recommend kl-f8-anime2.vae

  • CLIP skip = 1

  • Use Hires.fix to get higher quality image

    • Upscaler

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

    • Denoising strength 0.50~0.65

LoRA training info

• trained on sd-scripts by kohya_ss.

  • LyCORIS by KohakuBlueleaf

  • Thank you all a lot!

• based on Animefull-pruned model

• 86 images consist of

  • 4x low quality 34 images

  • 8x medium quality 21 images

  • 16x high quality 31 images

  • => 1 epoch = 800 images

• replaced character feature tags with narita top road \(umamusume\)

  • 1girl, multicolored eyes, bangs, horse tail, simple background, brown eyes, horse girl, yellow eyes, tail, symbol in eye, orange eyes, parted bangs, purple eyes, blonde hair, pink hair, medium hair, solo, long hair, multicolored hair, streaked hair, short hair, animal ears, white hair, horse ears, pink eyes, light brown hair, red eyes, brown hair

  • I think these tags are the essentials of Narita Top Road or

  • unnecessary for training because of duplication.

• resolution : 768x768 with aspect ratio bucketing.

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

And, special thanks to my best friend. DH.

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

IF 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 WORKING ON A WAY TO FIX THIS USING A CUSTOM NODE.

What's this LoRA

This is a LoRA trained on AI generated pictures with only two colors (made using the Bing Image Creator). 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

• https://civitai.com/models/33707

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

The LoRA is trained on NSFW pictures.

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 a custom LoRA or picture? Check out my commission page:

LoRA and Picture Commissions

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

Minami (ミナミ) Style LoHa

An Artstyle LoHa

With a "give it a try" attitude, I trained this model, even though its image training set was very small (around 80 images). However, the results were unexpectedly good. Below is a comparison of the results with and without the use of the model.

我抱着试一试的心态训练了这个模型，因为这个模型的图片训练集非常小 （大概只有80张左右）。但是让我出乎意料的是效果意外的好。下面是有或者没有使用模型的效果对比图。

Comparison W/O LoHa (对比图):

1girl, solo,upper body, looking at viewer, blue hair, simple background, chromatic aberration,dress, portrait, two-tone hair, long hair, <lora:mina_loha_v1a:1>
Negative prompt: verybadimagenegative_v1.3, EasyNegative
Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 5.5, Seed: 1251902067, Size: 512x1024, Model hash: c870e984cd, Model: anythingv5.5, Clip skip: 2, ENSD: 31337

1girl, anemone (flower), blue eyes, blue flower, blunt bangs, bob cut, bouquet, bug, butterfly, caterpillar, cherry blossoms, choker, cornflower, covered mouth, daffodil, dragonfly, dress, eyebrows hidden by hair, fern, flower, flower choker, flower dress, forget-me-not (flower), gloriosa (flower), holding, insect, leaf, lily (flower), long hair, long sleeves, looking at viewer, lying, magnolia, on back, oncidium, orange flower, orchid, pink flower, purple flower, solo, tulip, white dress, white flower, yellow flower, yellow tulip, white background, bangs, white hair, cowboy shot, frills, short hair, covering mouth, simple background, plant, arm up, <lora:mina_loha_v1a:1>
Negative prompt: verybadimagenegative_v1.3, EasyNegative
Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 5.5, Seed: 2003220760, Size: 768x768, Model hash: c870e984cd, Model: anythingv5.5, Clip skip: 2, ENSD: 31337

• Recommeneded prompts: long hair,flower,hair flower,white hair,upper body,earrings,bird,white dress,jewelry,long sleeves (Please check my sample pics for more info.)

• Trigger word: you don't need any trigger word for this LoHa.

• Recommeneded Weight: 0.7-1 (You can adjust it by your perference.)

• CFG scale: 5.5-7

• Check point recommendation: Anything V5

• How to use LoCon/LyCORIS? Please read instruction of a1111-sd-webui-locon.

If you like my work,please give me a like <3.

如果你喜欢我的模型请给我一个赞 <3.

もし、私のモデルが気に入ったら、「いいね！」をクリックしてください<3。

This is some momoko lycoris, I hope it is functional.

Personally I use them in combination.

It is the first time to make Lora, and I set the parameters by the way. I hope someone can provide a better configuration file. 500 datasets.

The displayed images were not carefully selected.

I originally intended to create a Lora-style artwork normally, but unexpectedly, the result turned out to be very good.

I tested with AnyLora and Anythingv4.5, with a weight of 0.6, Hires. Fix  Denoising 0.4.

The success rate is quite high, roughly estimated to be at least 80% of the generated images can be used directly. It also works well for multiple people situations with fewer limb errors.

Feel free to try it out.

展示图片未经精心挑选。

好久没有这么让我兴奋的lora了。本来想正常地做一个画风lora，没想到效果意外地好。

我使用anylora和anythingv4.5测试，权重0.6作用，Hires. Fix，Denoising0.4。成图率相当高，粗略估计至少有80%都可以直接用。对肢体错误少、多人情况下也不错。

推荐尝试！

To generate her star-shape pupil, please inpant with prompt 'rubypupil, eye close-up' and select 'Only masked'.

blue archive

Lemmtear Theta (レムティアシータ/彩姫 くるみ) from 魔法戦士レムティアナイツ～光の乙女たち～

I'm publishing this to celebrate of I finally reached 10K followers on Pixiv! Training Lemmtear's Lora is really a hassle. Why is it so hard to find CG of her wearing clothes in old style? I've trained this model four times already, and now I'm using the locon version, which works amazingly well.

苦逼跑图人在p站终于有1万粉丝了，发出来庆祝一下喵

Lemmtear算是三角社老作品里我特别喜欢的人设之一了，白丝大腿prpr。不过这些老gal的角色lora是真难练啊，穿衣服的cg太少了……前前后后炼了4次，感谢琥珀老师的lyCORIS救我狗命

Please follow my pixiv page for more example works: https://www.pixiv.net/users/23184632

0.8, 0.9 and 1

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。

CC License

A LYCORIS Locon of Mitama Yakmo from the spin-off Madoka series Magia Record: Puella Magi Madoka☆Magica Side Story.

You will have to install this extension made by KohakuBlueLeaf into your webui to fully use a LyCORIS Locon

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

This Locon was made with a collection of fanart images and a few screencaps from the gacha mobile game, she can be easily changed into multiple outfits and hairstyles as long as if you up the strength on the specific hairstyle and outfit you want to put her in (1.2-1.4 should be good enough) but yes you should always do your best to make Yui Horie voiced characters look pretty/cute in alternative outfits, that's the best thing about AI tbh

Compatibility

Trained on the NAI model, so it's pretty much compatible with every 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

normitama will be the token needed for her normal look, the additional tags needed to easily get her basic look will be

"long hair, single hair bun, ponytail, black ascot, bangs, blue eyes, blue skirt, large breasts, frills, grey hair, hair intakes, hair ornament, sleeveless, white gloves, blue tailcoat, blue vest, hair flower, blue rose, frills, black footwear"

you'll get her main look, the only thing that might not appear is the tailcoat or exact footwear but you'll get it with every batch so just keep on outputting, honestly not sure what's the reason for the tailcoat having RNG but it's probably because the training process and random cropping missed the few fanart pictures that had her fully showing off her outfit and tailcoat

mizumitama will be the token you use for her swimsuit outfit, the additional tags needed to generate this look will be

"large breasts, blue one-piece swimsuit, jewelry, blue eyes, blue rose, hair ornament, very long hair, o-ring, bracelet, necklace, braid, ring, grey hair, bangs, single braid, side braid, hair intakes"

newyearsmitama will be the token you use for her new years outfit, sadly not a lot of images were available to train this outfit so it isn't a 100% exact look to her new years costume but with the trigger token and additional tags you may get outputs that look close to it

"bangs, black footwear, black ribbon, blue eyes, blue kimono, blue rose, high heel boots, braid, floral print, frilled sleeves, frills, fur collar, fur trim, grey hair, hair flower, hair ribbon, long hair, long sleeves, obi, print kimono, sash, sidelocks, single side bun, white gloves, wide sleeves"

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/

有一点画风过拟，但是因为这个版本动作表现比较稳定，所以还是选择了这个epoch，记得开hiresfix消除画风过拟合的影响。我的例图权重为1。下面是示例prompt：

There is a bit of over-fitting in the style of painting, but because the action performance of this version is relatively stable, I still chose this epoch. Remember to enable hiresfix to eliminate the influence of over-fitting in the style of painting. My example graph has a weight of 1. Here is example prompt:

1girl,1boy,mating press, completely nude,nude male, sex,vaginal,leg lock,hug,kiss

请自己尝试替换一些tag，在此不再赘述

Please try to replace some tags by yourself, so I won’t go into details here

这种紧密接触的姿势很容易产生肢体错误，所以我建议善用inpaint功能

较低Denoising strength的i2i也可以基本消除画风影响（不要忘记取消lora）

This close contact pose is prone to limb errors, so I recommend making good use of the inpaint function

i2i with lower Denoising strength can also basically eliminate the influence of painting style(don't forget to cancel lora)

• 768*768 or similar resolution is recommended.

Yuisis from Granblue Fantasy.

Since this is the first time I created it, there may be many unnatural parts.

Model : AnyLoRA

Prompt : gbfyuisis