This is one of my first Lora's I have made, hope you all enjoy!
Settsett(league of legends)solo
Draw Stylized Ruins
cloud, cloudy_sky, day, sky, blue_sky, scenery, mountain, flower, gras
Lora trained on 168 images of feet facing viewer while lying.
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.
Toes are kinda random, chance is somewhat slim...
Prompts:
<lyco:feetaway1-000006:1.0>, on bed, indoors, simple background, feet, ass,
Trained on NAI, gen'd on AOM2 hard.
, on bed, indoors, simple background, feet, ass, hornet,twintails,boknd hair,green eyes,cowboy_hat,bikini,black_legwear,cape,thigh_boots,belt,hotpants
lumine_genshin_impact
N°20 Release, 素裳 - Sushang from Honkai: Star Rail - 崩壊スターレイル.
Put everything in Filter for seeing all images!
Click on Show more for Weight & Trigger recommendation, Also click on About This Version at Right Corner for more Info.
Well, it's was a challenge & It's was a Request, Well Sushang, Since she's has Very very low Art & IA Art on It, it was a good challenge (Like 14 Images at max in Booru + Pixiv).
But! i did manage & Trained The Lora to get 69+ Images, How ? Simple, i just upscaled, cropped, and flipped the angles of any images & I mixed a little with Game Screencaps at the best quality possible since i upscaled them with 4x-AnimeSharp, i tryed to be the most precise possible on Sushang, i think it will be hard to do better at the moment since Art on it is Very Very Low, But honestly, I'm quite satisfied about the result.
AnimeFullFinalPrunedFP16 used, so he work with the majority of Anime Model.
Flexibility is Ok as you can see in Examples images, best is to put 0.6-0.8 Weight for Flexibility.
About Trigger Words & Prompts: SilverWolfV4 (For Trigger) & for prompts just refer to examples image.
Weight recommended: 0.6-1. - Work at Weight 1, Best result: 0.7-1
Negative Embedding used: EasyNegative, BadPrompt2 and sometime negative_hand-neg.
Leave a Review & Share you're image in comments if you appreciate this Work :).
SushangV4
goku blackblack hairblack eyes
Based on AnythingV5.
lan 1 (tower of fantasy)mask on headkimono
Would appreciate images, feedback, requests, reviews, and of course comments in the comments. Please check out my other models if you like this one.
(this version is only doggystyle, since I tried other variations and it made it less stable, the everything below is just a description of the tag)
I HIGHLY RECOMEND NOT USING THIS MODEL FOR POV OR POSITIONS BESIDES DOGGYSTYLE AS IT IS NOT WORTH THE HASSLE
A Lora for adding a ripple effect to an ass. I will try and work on this, but I'm not sure how I could improve it. Hopefully someone sees this and makes a better version.
IMO ass ripple makes any hentai image better (unless it doesn't involve an ass), and like many annoying tags, people don't tag it.
weights 0.8 - 1.3
recommended tags: doggystyle(and its related tags).
optional tags: rough_sex, motion_lines, trembling, ^^^, (my go to when trying to simulate motion).
highres fix may help or make it worse.
changing the Lora's place in the prompt can greatly affect to output. ( i usually put it at the beginning then put it at the end and see which i like better, haven't messed with the middle)
This might work for other positions or pov, but through my testing it is extremely rare. (increasing the Lora to 1.3 helps with this). It seems that ass_grabs and the ass in general usually looks better with this Lora (though that may just be me)
AssRipple
Made from 32 screenshots of her 3d model (3 processed through IMG2IMG to make the face nicer) and 19 images made using an older version of this LoRA and various checkpoint models, to try and offset the 3d model look. Not happy with the result, but I'm busy, so I dunno if I'll try redoing it later or not.
参考データは3dモデルのスクショ32枚（顔を綺麗にするため、3枚はIMG2IMGで処理した）と、3dモデルのスタイルから離れるために、このLoRAの古い版といろんな別スタイルで作った19枚の絵でした。この結果に満足してないけど、忙しいからいつかやり直すかわかりません。
Recommended prompts （おすすめタグ）:
short hair, medium hair, jewelry, gem, green and white dress, off-shoulder, bow, white pantyhose, vertical-striped pantyhose, puffy sleeves, blunt bangs, detached collar, maid headdress, red bow, green collar, green shoes, corset, blue eyes, black hair
Recommended weight （おすすめウエイト）:
0.7-1.0
Even at these weights, the details tend to be slightly off depending on the checkpoint model used. You'll also need to use the recommended prompts.
このウエイトでも、使ってるチェックポイントモデルによって詳細がよく少しずれてます。おすすめタグも使う必要です。
From my experience, LoRAs seem to work by overwriting what the AI originally intended to draw, based on your prompts. So the more complex a character is, the more you'll need to explain the character's basic appearance. For example, if the AI intended to draw a tight swimsuit, and your LoRA is a big dress, it'll probably make a swimsuit that looks like your LoRA dress (ie. with the skirt missing).
今までの個人体験ですが、LoRAは使ってるタグによってAIが元々描くつもりのキャラクタを書き換えてるっぽいです。ということは、キャラのデザインがもっと複雑場合、その外見の基本をもっとはっきり説明する必要だと思います。例：元々キツイ水着を描くつもりで、使ってるLoRAは大きなドレス。結果は多分LoRAのドレスに似てる水着（スカートなし）の絵になります。
Problems (問題点):
Patterns and minor details are often slightly off.
パターンと細かい詳細がよく少しずれてます。
At 0.8-1.0 the style of the face will be less influenced by the checkpoint model.
0.8-1.0の場合、顔にチェックポイントモデルからの影響が減ります。
Models used for screenshots （スクショに使ったモデル）:
Counterfeit v3.0
Based65 Final Mix
Tsubaki v1.0
CardosAnime v1.0
ExpMixLine v2
Meina Mix v9
SweetMix v1.6/1.7
CoffeeMix v2.0
Anything v4.5 (with Hoyo Artstyle LoRA v2)
使用建议:
配合一些Lora, TI, VAE等建筑使用文件
如果你觉得不错，就给我打个五星评价吧
如果您有任何问题，请通过xsarchitecturalstudio@gmail.com与我联系
Suggestions for use:
Lora, TI, VAE and other documents for indoor use
If you think it's good, give me a five-star review
If you have any questions, please contact me at xsarchitecturalstudio@gmail.com
XS轻量模型系列在库编号100+轻量模型将有新的工作流和使用方式，大致分为主体轻量模型与辅助轻量模型（LoRa、Loha、Perfusion、TI、VAE等）
主体轻量模型：XSArchi_1***V1，V1将作为主体轻量模型编号进行规整命名，在工作流中起到主导图像生成的工作，复合功能将决定于轻量模型功能。使用权重在0.5-1.3左右。
辅助轻量模型：XSArchi_1***V1.1，V1.1（即此Lora第二版）将作为辅助轻量模型编号进行规整命名，在工作流中起到辅助图像生成的工作，无复合功能，分为视角、风格、应用场景等20余种功能轻量模型，详细功能将在轻量模型介绍中标出。使用权重在0.1-0.5左右。
V2+轻量模型将被更新或迭代。
新轻量模型将分为两种不同的训练、融合、提取等方式进行制作，在库编号相同但版本不同的轻量模型，在训练集的处理上大致无异。在库编号不同但版本相同的轻量模型，功能不同外，训练集及训练方式也各不相同。
新版本将加入中英文命名的方式，解决大家对于模型标识、分类出现的困难。
其余版本介绍：
在库编号100-的轻量模型为早期试用模型，将被迭代。
在库编号100+的轻量模型为正式模型，分为商用及非商用模型，将会持续更新。
在库编号百位数以复数形式出现的为独立授权模型，针对不同平台、企业、组织、个人等进行授权、商用、非商用流通。
在库含有编号的轻量模型被认作为垂类模型库在库模型，将配合XS工作流进行使用。
XS Lightweight model series in the library number 100+ lightweight model will have a new workflow and use methods, roughly divided into the principal lightweight model and auxiliary lightweight model (LoRa, Loha, TI, VAE, etc.).
Body lightweight model: XSArchi_1***V1, V1 will be structured and named as the body lightweight model number, and play a leading role in image generation in workflow. Composite function will be determined by the function of the lightweight model. The right to use weight is around 0.5-1.3.
Auxiliary lightweight model: XSArchi_1***V1.1, V1.1 (i.e., the second version of Lora) will be structured and named as the auxiliary lightweight model number, and play an auxiliary role in image generation in workflow. There is no compound function, and it is divided into more than 20 functional lightweight models such as perspective, style and application scene. The right to use is about 0.1 to 0.5.
V2+ lightweight model will be updated or iterated.
The new lightweight model will be divided into two different ways of training, fusion, extraction and so on. The lightweight model with the same library number but different versions will be roughly the same in the processing of the training set. In addition to the different functions of lightweight models with different library numbers but the same version, the training sets and training methods are also different.
The new version will add the way of naming in Chinese and English to solve the difficulties of model identification and classification.
Introduction to other versions:
The lightweight model in library number 100- is an early trial model that will be iterated.
The lightweight models in the library number 100+ are official models, divided into commercial and non-commercial models, and will be updated continuously.
The number of hundreds in the database appears in the plural form as an independent authorization model, which carries out authorization, commercial and non-commercial circulation for different platforms, enterprises, organizations, individuals, etc.
The numbered lightweight models in the library are considered as vertical model models and will be used in conjunction with the XS workflow.
sometimes we release beta versions to community testing before upload to Civitai
Weight (Strength) used on images: 1
Dim: 8
Alpha: 4
150 images with flipped copies
📌Training model：AnythingV5V3_v5PrtRE.safetensors
📌LoRA set：0.8
📌Prompt：Yamada Anna, white collared shirt, partially unbuttoned(部分解開), turtleneck(高領毛衣), black pleated skirt(黑打褶裙), plaid dress(格子連衣裙), checkered skirt(格子半身裙), plaid skirt(花格裙), suspender skirt(吊帶裙), suspenders(吊帶褲), blue jacket, school uniform, open clothes, long sleeves, ponytail, black lagwear, kneehighs, white footwear, uwabaki(學生室內鞋), blazer
📢I release my models for free, if you like my works, you can support me and buy me a drink.☕☕☕https://www.buymeacoffee.com/hansom283
Yamada Anna