PixAI Membership
PixAI
PixAI
Generate
Sign in
Home
Generation
Models
Studio
NEW
Contest
Create with Mio.2
AI Tools
NEW
Guide

SDXL

One obsession

Create

  • AI Image Generator
  • AI Animation Generator
  • AI Tools
  • Thematic generators
  • Train LoRA
  • Mio.2 Agent
  • Studio

Discover

  • Trending Tags
  • Ranking
  • Model Market
  • Contest
  • News

About

  • Guide
  • PixAI Blog
  • Tsubaki.2
  • Meet Mio, Our Mascot
  • Content Rules

Pricing & Help

  • Membership
  • Buy Credits
  • Contact

Mobile App

  • Get the PixAI app
  • App Store
  • Google Play
© 2026 PixAI
  • Terms of Service
  • Privacy Policy
  • Copyright Policy
  • Attributions
  • Specified Commercial Transactions Law
One obsession - AI Model cover image
Uploaded at
Jun 14, 2025, 3:25 PM
Uses
228k
Reviews
Excellent  (12)
Sampling Steps

30

Sampling Method

Euler a

CFG Scale

2.5

Negative

worst quality,normal quality,anatomical nonsense,bad anatomy,interlocked fingers,extra fingers,watermark,simple background,transparent,low quality,logo,text,signature,face backlighting,

Permissions
  • Allow image generation & sharing
  • Allow users to download this model
  • Allow commercial use of generated images

Recommended LoRAs

DMD2 - AI Model cover image

User LoRA

DMD2
6.31k
Anime Enchancer XL-v5 - AI Model cover image

User LoRA

Anime Enchancer XL-v5
4.03k
Hands XL - AI Model cover image

User LoRA

Hands XL
1.33k
Background Enhancer [SDXL] - AI Model cover image

User LoRA

Background Enhancer [SDXL]
6.69k

Description

If you like my work, please give me a good review and click the thumbs-up icon. Please give me more feedback! It's free, and it can keep me motivated to create.如果你喜欢我的作品,请给我一个好评并点击大拇指图标。多多返图!这是免费的,而且能让我保持创作动力。Still, I repeat, the new version may not necessarily be the best. Experience it yourself and have fun!^^!绘画标签都藏在图片提示词中!寻找画师串也是乐趣所在!例图未加任何lora,其他自己探索吧!还是那句话,各有特色新版本不一定是最好的。自己体验,玩得开心!AI绘画QQ交流群:4654350 Discord:https://discord.gg/7RNWedD59H^^! The painting tags are all hidden in the picture hints! Finding artists to collaborate with is also a lot of fun! The example picture hasn't been enhanced with any lora, and the rest is up to you to explore! As I've said before, the new version may not be the best. Experience it yourself and have fun! Discord:https://discord.gg/7RNWedD59H推荐一个forge或webui的插件,方便 Illustrious的动漫角色的选择,插件可以扩展URL安装,插件链接如下:https://github.com/lanner0403/WAI-NSFW-illustrious-character-select可盐可甜可干爽可油腻可3D可2D 可涩涩可清纯 打造泛用性最强二次元Illustrious微调大模型!(狗头)光影方面建议加:(profile light,contour light), 等来强化光影如果希望画面变暗可加: (dark,chiaroscuro,hight-contrast:1.7),人物半写实可加:((realistic, photorealistic, realistic shading)),还原动漫可以加:(anime coloring ),(anime screenshot)动漫上色,截图来还原★NOW★1.4,这个版本变化有点大,默认是2.5D的人物,(可以通过画风标签的tag改变)比上个版本加强了一点3D,画面整体干净,质感不错,其他轻微调整。1.4 This version has quite a lot of changes. The default characters are 2.5D, (which can be changed through the tag of the art style) and it has been slightly enhanced in 3D compared to the previous version. The overall image is clean, and the texture is quite good. There are other minor adjustments.★One obsession Illustrious2.0 1.1★1.3版本,动漫人物还原出色;优化了手脚的稳定性和少许改变,优化光影,优化构图,改善了丝袜的质感;2D和2.5D都有很好的表现。重要的事情再说一遍,例图未加任何Lora。Version 1.3, The anime characters are well-reproduced; the stability of hands and feet has been optimized, and there are some minor changes. The light and shadow have been optimized, as well as the composition. The texture of the stockings has been improved; both 2D and 2.5D have a good performance. Let me repeat the important thing: the example pictures have not been enhanced with any Lora.1.2版本:我们的口号是不加Lora也很好看!Euler画风偏漫画一些,Euler a画风整洁一些(整体差别不大)。建议CFG3-6低一些,我喜欢CFG3.5,20-30steps,不过用绘画标签依然能展现2D效果。偏写实了一点点,还原各个年龄的角色,极致光影回归,增强更华丽背景,人物再次增强质感强化了2.5D,强化肢体动作,眼睛更迷人,增加了可爱的数据集。Version 1.2: Our slogan is that it looks great even without Lora! Euler's art style is more like a comic, while Euler a's art style is more neat(The overall difference is not big). We recommend setting CFG3-6 lower, I prefer CFG3.5, 20-30 steps. However, using painting labels can still show the 2D effect. It's a bit more realistic, restoring characters of various ages, and returning to the extreme light and shadow. The background is more gorgeous, and the characters are once again enhanced in texture, strengthening 2.5D, and enhancing the body movements. The eyes are more charming, and a cute dataset that is more adorable has been added.1.1版本:较1.0版本增强了人物和场景的细节和3D化,光影和多人肢体增强,颜色更加鲜艳。:)没办法,我总是在这几个问题来回横跳。^^!绘画标签都藏在图片提示词中!寻找画师串也是乐趣所在!例图未加任何lora,其他自己探索吧!the details and 3D rendering of characters and scenes, improved the lighting and multiplayer limbs, and made the colors more vibrant. :) I can't help it, I'm always jumping back and forth between these few issues. ^^! The painting tags are all hidden in the picture prompts! Finding artists to collaborate with is also a lot of fun! The example images have not been enhanced with any lora, and the rest is up to you to explore!1.0版本,终于更新到Illustrious2.0了,和08.09版本一样已经支持了简单的自然语言,还是2.3D画质,兼容各种画风标签和lora,泛化性不错,继承了09版本的优点黑色场景很纯正,比09版本画面光影柔和一些,没有09版本场景细节丰富,所以干净一些,有利有弊吧。绘画标签都藏在图片提示词中!寻找画师串也是乐趣所在!例图未加任何lora,其他自己探索吧!A new beginning, thank you for everyone's liking. The 1.0 version has finally been updated to Illustrious2.0, which, like the 08.09 version, already supports simple natural language. It still has 2.3D quality, and is compatible with various art styles and lora. Its generalization is quite good, inheriting the advantages of the 09 version. The black scenes are very pure, and the light and shadow in the pictures are softer than those in the 09 version. However, the scene details are not as rich as those in the 09 version, so it's cleaner. There are advantages and disadvantages to it. The painting tags are all hidden in the picture prompts! Finding artists to follow is also a lot of fun! The example pictures have not added any lora, and you can explore the rest by yourself!09版本,有较大的改变。光影俱佳,有着不输V-pred的表现,增强暗部表现,颜色调整,另外我觉得这个版本眼睛最好看。增强了人物的质感,延续干净的风格,背景细节丰富,个人推荐设置CFG3.5我比较喜欢,增强了3D感,自定义2.35D吧,丝袜一如既往的优秀!例图未加任何lora,其他自己探索吧!The 09 version has undergone significant changes. The lighting and shadows are excellent, and its performance is not inferior to V-pred. It has improved the performance of dark areas and color adjustment. In addition, I think the eyes look the best in this version. It has enhanced the texture of the characters and maintained a clean style. The background details are rich. Personally, I recommend CFG3.5, which I prefer. It has enhanced the 3D effect, and the custom 2.35D is as excellent as ever! The example images have not been enhanced with any lora. Explore more on your own!★One obsession Rouwei★08版本使用了最新的illustrious微调模型rouwei的最新数据集,兼容自然语言,出图有新鲜感,例图未加任何lora,恢复之前版本1-5好评的皮肤红润,人物偏可爱风,一如既往的整体干净画风,其他自己探索吧!The 08 version uses the latest Illustrious fine-tuning model and Rouwei's latest dataset, which is compatible with natural language. The output images look refreshing, and the example images have not been enhanced with any Lora. It restores the previous versions 1-5's praised skin redness, and the characters are more cute. As always, the overall style is clean. You can explore the rest by yourself!★One obsession noobAI★07版本,这个版本是noobai,集大成之作,2D,3DCG,动漫,半写实都有很好的表现,泛用性很好;颜色调整为最佳,还原度高特别舒服,不会饱和;画面干净且复杂,场景会有更多细节,肢体很好,例图均为直出,打开ADetailer会加强脸部细化。Version 07, this version is a masterpiece, featuring excellent performances in 2D, 3DCG, anime, and semi-realistic styles. It's very versatile. The color adjustment is optimal, with a high degree of restoration, which is particularly comfortable and not saturated. The images are clean and complex, and the scenes will have more details. The body shapes are excellent, and the example images are all taken directly. Opening ADetailer will enhance the facial details.★One obsession Illustrious1.0★05版本,这个版本比较中和,强化手脚,告别默认娃娃脸,幼化成熟都可以,提升质感,详细看例图吧。Version 05: This version is quite balanced, enhancing the hands and feet, and getting rid of the default baby face. It can both make you look younger and more mature, and improve the texture. Please check the detailed examples.04版本,回归2.2D干净动漫风格,偏萝莉可爱简洁型。出动画的时候,加上“anime screenshot,1990s (style),”等自己添加画师串使 其复杂吧。Version 04, a return to the clean 2.2D anime style, leaning towards a cute and simple lolita style. When releasing the animation, add "anime coloring, anime screenshot," and let the artists make it more complex.★One obsession Realistic★03版本,这是一个2.5D的半写实动漫大模型,它擅长描述成熟女性。但萝莉也可以出的,大萝莉,不输入画师串干扰的话整体偏真实一些。Version 03:This is a 2.5D semi-realistic anime model, which is good at depicting mature women. But it can also create lolis, even big lolis. Without the interference of the artist's personal style, the overall look is quite realistic.★One obsession Color style★02版本,有着极致的光影,做了减法它并不能代替01版本,强化了肤色的质感,强化了黑白丝袜的质感,减弱了风格使画面简洁一些,没有01版本画面色彩丰富,更好的兼容lora,强化了光影和手脚。The 02 version has an extreme light and shadow, and even though it has been simplified, it still can't replace the 01 version. It has weakened the style and made the picture simpler. Version 02, which is more compatible with Lora, has improved lighting and shadows as well as hands and feet.01版本,有着迷人的色彩,加强了风格化动漫,丰富的背景。V1.0, Enhance the stylized processing, extreme colors, and rich details in the images.You can use ADetailer to enhance facial details:https://github.com/Bing-su/adetailer可以使用 ADetailer 来增强面部细节:https://github.com/Bing-su/adetailerEntering "realistic, photorealistic" will give you different styles.提示词里面加入"realistic, photorealistic,"萝莉变御姐!大图脸小的话就开AD修脸吧!放大会有更多细节!一般画师串都兼容,如果想写实一些就加入提示词, "((realistic, photorealistic, realistic shading)), "可以加权重两个或三个括号,大家玩的愉快吧推荐我喜欢的分辨率:768X1280 768X1344 832X1216 896X1152Model Recommended Resolution:768X1280 768X1344 832X1216 896X1152Illustrious 1.0(noobAI):896 x 1152、1216 x 832、832 x 1216、1344 x 768、768 x 1344、1536 x 640、640 x 1536,Illustrious 2.0{Rouwei Illustrious1.1}支持更大分辨率:1536x1536、1728x1344、1344x1728、1824x1248、1248x1824、2016x1152、1152x2016正面词 Positive Prompt:masterpiece, best quality, amazing quality, very awa,absurdres,newest,very aesthetic,depth of field, highres,反义词 Negative Prompt:worst quality,normal quality,anatomical nonsense,bad anatomy,interlocked fingers,extra fingers,watermark,simple background,transparent,low quality,logo,text,signature,face backlighting,Recommended settings:Steps: 25-35CFG scale: 3-6Sampler: Euler a Euler

Comments1