Quick Generate
Select a style, write what you want to see and click Generate
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,

,falling_above, (1girl:1.3), solo,<lora:falling_above-10:0.5>   ,(falling:1.3),(midair:1.2),floating,(from above:1.2), tree,forest,lake,autumn,floating clothes,



Steps: 38, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 2764840952, Size: 1152x1728, Model hash: e4a30e4607, Model: favorite_majicmixRealistic_v6, Denoising strength: 0.54, Clip skip: 2, Tiled Diffusion upscaler: R-ESRGAN 4x+, Tiled Diffusion scale factor: 1.5, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Tile tile width': 96, 'Tile tile height': 96, 'Tile Overlap': 48, 'Tile batch size': 8, 'Upscaler': 'R-ESRGAN 4x+', 'Upscale factor': 1.5, 'Keep input size': True}", ControlNet 0: "preprocessor: none, model: control_v11f1e_sd15_tile [a371b31b], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: ControlNet is more important, preprocessor params: (512, 64, 64)"
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,
,falling_above, 1girl, solo,   ,falling,midair,floating,roof,from above, floating clothes,mountain,rock,cliff,water,
What you want in the image.
Clear
Random

Model Description by Creator

Please read the following to improve quality

  • 尽情发挥你的想象力吧!她,到底可以在哪里坠落呢?喜欢我模型的朋友,想要支持我的话,请到我的哩布模型页面找到一个你喜欢的模型在其中发表图片。非常感谢!

坠落是一种无力感,一种已知结局却无法改变的绝望。

falling_blur:

  • 建议权重:0.4~0.5可能多余肢体,但更灵活,换衣服,换姿势。0.5~0.6基本姿势固定,想要完全符合的肢体就调大权重,背景会更丰富。

  • 建议正向提示词:falling,(from below:1.2),falling_blur,depth of field,floating,focused, 其他想写什么就写什么。可以多试试。

今天是端午节,祝大家端午安康。做了一整天falling的新版本,该版本考虑到了天空占比问题以及景深。并且训练集扩大了三倍,每张图片都使用tile模型放大过。效果很可观。降低权重、不使用触发词应该可以只得到风景(但在一个小时的测试中忘记试了),赛博朋克,超现实等应该都可以生成。

现在可以脱掉衣服了😉

在端午这一天从早上八点搞到晚上八点。

以下为【致郁】内容,敏感者请勿阅读。写在这里只是自我抒发情感而已,因为在其他地方根本没人会看。

我的大学可以说是完全不务正业的。大一上半死不活,大一下开始搞玄学(真的玄学,不是比喻)。。搞到大三上,被迫学了一学期专业课,大三下又开始沉迷AI绘画。不知道是不适应大学还是怎么回事。我唯一对父母承诺过的就只是活着,没有朋友,没有爱的人,人类的情感逐渐从我身上脱离了。对我而言,没有未来这种东西,只有此时此刻,只有此时此刻我所感受到的,我所经历的,我所体验的,我所品尝的。这些东西我之前不愿意给别人说,因为他们都说要离我这样的人远一点。我自己也不喜欢与人亲近,天生如此吧。前两天一整天不吃饭也不饿,只是有点胃痛而已。晚上总是睡不好,今天凌晨一点睡,三点就醒了,然后睡不着了。

我真的不知道我想要的到底是什么,我应该做什么,我应该去哪里,我只是一个自我中心的普通人类罢了,几乎没有什么能让我快乐了。只是我想到你们直接用我做的模型,会觉得有一点点幸福罢了。原本我是学理工的,但是看到”美“这种东西才真正能让我平静下来。我需要的是创造。请不要评论此内容,我已经没救了。

以下是感谢名单:

【支持者:暂无】

【帮助者:暂无】

Falling is a feeling of powerlessness, a desperation with a known end that cannot be changed.

falling_blur:

Recommended weight: 0.4~0.5 May have extra limbs, but more flexible, change clothes, change posture. 0.5~0.6 The basic posture is fixed, and if you want to fully match the limbs, increase the weight, and the background will be richer.

Recommended positive prompt words: falling,(from below:1.2),falling_blur,depth of field,floating,focused,

Whatever else you want to write. You can try more.

Today is the Dragon Boat Festival. I wish you all good health. A new version of falling has been made all day, which takes into account the sky proportion problem and depth of field. And the training set is three times larger, with each image magnified using the tile model. The results are impressive. Lowering the weights and not using trigger words should result in just scenery (but forgot to try it in an hour of testing), cyberpunk, surreal, etc., should all be generated.

You can take off her clothes now 😉

I work from 8 am to 8 pm on the Dragon Boat Festival day...

Here is the list of thanks:

【Supporter: None for now】

【Helper: None for now】

falling_above and falling_side:

自由落体,两个型号,两种视角。

因为之前尝试了无数次,想画坠落,用各种controlnet模型都做不到,出现混乱的手脚。原因也可能是我英语的描述能力不好。反正做了这个,也试了各种方法。只能退而求其次了。

  • 注意用图生图放大改善面部,推荐用multidiffusion放大和controlnet tile模型,加大重绘幅度,选择“controlnet更重要”模式

  • 注意from_side型号面部无法识别的时候可以下载图片旋转180°再图生图。

建议权重在0.9左右,不可过小,否则手脚混乱。

Falling is a sense of powerlessness, a despair of knowing the outcome but being unable to change it.

Free fall, two perspectives, two models.

Because I have tried countless times before to depict falling, with various controlnet models still can't, there were chaotic hands and feet. The reason may also be my poor ability to describe in English. Anyway, I have done this and tried various methods. I can only settle for second best.

  • Pay attention to using img2img to enhance facial features, recommended methods include multidiffusion and controlnet tile model, increasing the scale of redrawn image, and selecting the "controlnet is more important" mode.

  • Note that when the facial features of the "from_side" model cannot be recognized, you can download the image, rotate it 180°, and then generate the image.

I suggest using a weight around 0.9, not too small, as otherwise the actions will be chaotic.

Images Generated With This Model

(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,
,falling_side, 1girl,face,eyes, solo, <lora:falling_side-10:0.5> city,river,road,rooftop,
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres, ,falling_side, 1girl,face,eyes, solo, <lora:falling_side-10:0.5> city,river,road,rooftop,
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,
,falling_side, 1girl,face,eyes, solo, <lora:falling_side-10:0.5> city,river,road,rooftop,
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres, ,falling_side, 1girl,face,eyes, solo, <lora:falling_side-10:0.5> city,river,road,rooftop,
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,
falling_side, 1girl, solo, <lora:falling_side-10:1> ,forest, ocean, rock,rooftop,( Young Bronze Dragon:1.5)
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres, falling_side, 1girl, solo, <lora:falling_side-10:1> ,forest, ocean, rock,rooftop,( Young Bronze Dragon:1.5)
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,
falling_side, 1girl, solo, <lora:falling_side-10:1> ,forest, ocean, rock,rooftop,( Ancient Green Dragon:1.3)
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres, falling_side, 1girl, solo, <lora:falling_side-10:1> ,forest, ocean, rock,rooftop,( Ancient Green Dragon:1.3)
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,

,falling_above, (1girl:1.3), solo,<lora:falling_above-10:0.5>   ,(falling:1.3),(midair:1.2),floating,(from above:1.2), tree,forest,lake,autumn,floating clothes,



Steps: 38, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 2764840952, Size: 1152x1728, Model hash: e4a30e4607, Model: favorite_majicmixRealistic_v6, Denoising strength: 0.54, Clip skip: 2, Tiled Diffusion upscaler: R-ESRGAN 4x+, Tiled Diffusion scale factor: 1.5, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Tile tile width': 96, 'Tile tile height': 96, 'Tile Overlap': 48, 'Tile batch size': 8, 'Upscaler': 'R-ESRGAN 4x+', 'Upscale factor': 1.5, 'Keep input size': True}", ControlNet 0: "preprocessor: none, model: control_v11f1e_sd15_tile [a371b31b], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: ControlNet is more important, preprocessor params: (512, 64, 64)"
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres, ,falling_above, (1girl:1.3), solo,<lora:falling_above-10:0.5> ,(falling:1.3),(midair:1.2),floating,(from above:1.2), tree,forest,lake,autumn,floating clothes, Steps: 38, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 2764840952, Size: 1152x1728, Model hash: e4a30e4607, Model: favorite_majicmixRealistic_v6, Denoising strength: 0.54, Clip skip: 2, Tiled Diffusion upscaler: R-ESRGAN 4x+, Tiled Diffusion scale factor: 1.5, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Tile tile width': 96, 'Tile tile height': 96, 'Tile Overlap': 48, 'Tile batch size': 8, 'Upscaler': 'R-ESRGAN 4x+', 'Upscale factor': 1.5, 'Keep input size': True}", ControlNet 0: "preprocessor: none, model: control_v11f1e_sd15_tile [a371b31b], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: ControlNet is more important, preprocessor params: (512, 64, 64)"
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,
,falling_above, 1girl, solo,   ,falling,midair,floating,roof,from above, floating clothes,mountain,rock,cliff,water,
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres, ,falling_above, 1girl, solo, ,falling,midair,floating,roof,from above, floating clothes,mountain,rock,cliff,water,
A photo of No-one12 a woman dressed as cyberpunk woman,Dark brown eyes, portrait, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, cinematic lighting, 4K HQ, sharp focus, (Ultra realistic [[photo]], detailed face:1.0), (detailed eyes:1.0), (realistic photo:1.1), (masterpiece:1.0), detailed background, by Antonio J. Manzanedo
N/A
Hidden For Public Users
A photo of No-one12 a woman dressed as cyberpunk woman,Dark brown eyes, portrait, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, cinematic lighting, 4K HQ, sharp focus, (Ultra realistic [[photo]], detailed face:1.0), (detailed eyes:1.0), (realistic photo:1.1), (masterpiece:1.0), detailed background, by Antonio J. Manzanedo
A photo of No-one12 a woman dressed as scarlet witch,Dark brown eyes, portrait, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, cinematic lighting, 4K HQ, sharp focus, (Ultra realistic [[photo]], detailed face:1.0), (detailed eyes:1.0), (realistic photo:1.1), (masterpiece:1.0), detailed background, by Antonio J. Manzanedo
N/A
Hidden For Public Users
A photo of No-one12 a woman dressed as scarlet witch,Dark brown eyes, portrait, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, cinematic lighting, 4K HQ, sharp focus, (Ultra realistic [[photo]], detailed face:1.0), (detailed eyes:1.0), (realistic photo:1.1), (masterpiece:1.0), detailed background, by Antonio J. Manzanedo
A photo of No-one12 a woman dressed as scarlet witch,Dark brown eyes, portrait, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, cinematic lighting, 4K HQ, sharp focus, (Ultra realistic [[photo]], detailed face:1.0), (detailed eyes:1.0), (realistic photo:1.1), (masterpiece:1.0), detailed background, by Antonio J. Manzanedo
N/A
Hidden For Public Users
A photo of No-one12 a woman dressed as scarlet witch,Dark brown eyes, portrait, highly detailed, digital painting, artstation, concept art, smooth, sharp focus, illustration, cinematic lighting, 4K HQ, sharp focus, (Ultra realistic [[photo]], detailed face:1.0), (detailed eyes:1.0), (realistic photo:1.1), (masterpiece:1.0), detailed background, by Antonio J. Manzanedo
(masterpiece:1.2), (best quality:1.2), hyper realistic,highestres,absurdres,extremely detailed,volumetic,dramatic lighting,
1girl,detailed eyes,face,looking to the side,<lora:detailed_eye-10:1> 

Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 1274936605, Size: 2048x2048, Model hash: d819c8be6b, Model: majicmixRealistic_v4, Denoising strength: 0.35, Clip skip: 2, ENSD: 31774, Mask blur: 4, Tiled Diffusion upscaler: 4x-UltraSharp, Tiled Diffusion scale factor: 2, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Latent tile width': 96, 'Latent tile height': 96, 'Overlap': 48, 'Tile batch size': 8, 'Upscaler': '4x-UltraSharp', 'Scale factor': 2, 'Keep input size': True}"
N/A
Hidden For Public Users
(masterpiece:1.2), (best quality:1.2), hyper realistic,highestres,absurdres,extremely detailed,volumetic,dramatic lighting, 1girl,detailed eyes,face,looking to the side,<lora:detailed_eye-10:1> Steps: 20, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 1274936605, Size: 2048x2048, Model hash: d819c8be6b, Model: majicmixRealistic_v4, Denoising strength: 0.35, Clip skip: 2, ENSD: 31774, Mask blur: 4, Tiled Diffusion upscaler: 4x-UltraSharp, Tiled Diffusion scale factor: 2, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Latent tile width': 96, 'Latent tile height': 96, 'Overlap': 48, 'Tile batch size': 8, 'Upscaler': '4x-UltraSharp', 'Scale factor': 2, 'Keep input size': True}"
(photo of ohwx man:0.5), (((in ske style))), ((cel animation)), illustration, ((seinen, isekai, action, anime))
ControlNet 0: "preprocessor: openpose_full, model: control_v11p_sd15_openpose [cab727d4], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (512, 64, 64)", 
ControlNet 1: "preprocessor: canny, model: control_v11p_sd15_canny [d14c016b], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (512, 100, 200)"
N/A
Hidden For Public Users
(photo of ohwx man:0.5), (((in ske style))), ((cel animation)), illustration, ((seinen, isekai, action, anime)) ControlNet 0: "preprocessor: openpose_full, model: control_v11p_sd15_openpose [cab727d4], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (512, 64, 64)", ControlNet 1: "preprocessor: canny, model: control_v11p_sd15_canny [d14c016b], weight: 1, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: Balanced, preprocessor params: (512, 100, 200)"
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,solo focus, 

1girl,floating,on back,  (falling:1.3),falling_blur,<lora:falling_blur-10:0.42> ,(solo:1.3),(from below:1.2),midair,(depth of field:1.3),

<lora:lowRA:0.5> ,dark theme,evil,crush,dynamic,crazy,motion,(surreal:1.3)(,floating objects:1.3),books,surrending,susbending
Negative prompt: easynegative,bad-hands-5, lr,(monochrome:1.2),facial mark, facial tattoo,border,frame,(pantyhose),(low quality:1.3),bad artist's work,ground,(on ground:1.3),(2girls:1.3),
Steps: 24, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 3885471310, Size: 1024x1536, Model hash: cbfba64e66, Model: CounterfeitV30_v30, Denoising strength: 0.63, Clip skip: 2, Tiled Diffusion upscaler: R-ESRGAN 4x+ Anime6B, Tiled Diffusion scale factor: 2, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Tile tile width': 96, 'Tile tile height': 96, 'Tile Overlap': 48, 'Tile batch size': 8, 'Upscaler': 'R-ESRGAN 4x+ Anime6B', 'Upscale factor': 2, 'Keep input size': True}", ControlNet 0: "preprocessor: tile_resample, model: control_v11f1e_sd15_tile [a371b31b], weight: 0.85, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: ControlNet is more important, preprocessor params: (512, 1, 64)"
N/A
Hidden For Public Users
(masterpiece:1.3),(best quality:1.2),8k,absurdres,unity 8k wallpapper,(extremely detailed:1.3),highestres,solo focus, 1girl,floating,on back, (falling:1.3),falling_blur,<lora:falling_blur-10:0.42> ,(solo:1.3),(from below:1.2),midair,(depth of field:1.3), <lora:lowRA:0.5> ,dark theme,evil,crush,dynamic,crazy,motion,(surreal:1.3)(,floating objects:1.3),books,surrending,susbending Negative prompt: easynegative,bad-hands-5, lr,(monochrome:1.2),facial mark, facial tattoo,border,frame,(pantyhose),(low quality:1.3),bad artist's work,ground,(on ground:1.3),(2girls:1.3), Steps: 24, Sampler: DPM++ 2M Karras, CFG scale: 7, Seed: 3885471310, Size: 1024x1536, Model hash: cbfba64e66, Model: CounterfeitV30_v30, Denoising strength: 0.63, Clip skip: 2, Tiled Diffusion upscaler: R-ESRGAN 4x+ Anime6B, Tiled Diffusion scale factor: 2, Tiled Diffusion: "{'Method': 'MultiDiffusion', 'Tile tile width': 96, 'Tile tile height': 96, 'Tile Overlap': 48, 'Tile batch size': 8, 'Upscaler': 'R-ESRGAN 4x+ Anime6B', 'Upscale factor': 2, 'Keep input size': True}", ControlNet 0: "preprocessor: tile_resample, model: control_v11f1e_sd15_tile [a371b31b], weight: 0.85, starting/ending: (0, 1), resize mode: Crop and Resize, pixel perfect: True, control mode: ControlNet is more important, preprocessor params: (512, 1, 64)"

Train Models

Train AI Models On Your Own Pictures

You can use your own pictures to train AI models. Reimagine yourself in new ways with our AI-powered image generator.
Create pictures of yourself around the world, in your dream outfit, as a video game character, or in countless art styles. The possibilities are endless.
Train Your Own Model

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.