TextualInversions

31 year old Russian Instagrammer and nude model.

Paul Mirabel, born November 29, 1995 in Montpellier, is a French comedian.

A textual inversion embedding for generating dresses and tops made of rope.

Use Tips

Add r0p3dr3ss to your prompt.

A textual inversion when added to the prompt depending on the models its used with will give female models more muscles. Its fun when used with other Textual Inversion models at the same time. I only trained it with women, so doesn't work for men; You will get a buff women. This is my first try. May make it better if I know how to. Just was playing around and seeing what I get. I trained the model with buff women's bodies, I did not included the faces. so whatever face it give is totally the ai. I wanted it to be able to use the face from another IT. I usually will generate 10 images and 2 or 3 will be good when using with other TI models.

Textual Inversion embedding I did of actress Vanessa Hudgens. Used about 128 reference photos of her from photo shoots, walking, etc

I'm happy with how this one turned out. I dont think I needed to run it as long with as many images but I get some pretty great results with this so I guess I picked some good ones

Textual Inversion embedding I did of actress Kristen Bell. Used about 52 reference photos of her from photo shoots, walking, etc

Textual Inversion embedding I did of actress Kristen Bell. Used about 87 reference photos of her from photo shoots, walking, etc.

Might need to lower the image count in my next run. You still get the pretty good generations with this one.

Textual Inversion embedding I did of actress Hayden Panettiere (Heroes, Scream, Nashville). Used about 75 reference photos of her from photo shoots, walking, etc.

With the right prompts you can get some pretty life like visual showings of her. I could probably do a bit more tweaking on my photos but I think you can get good results with this one.

Lucy Katherine Pinder (born 20 December 1983) is a British glamour model and actress. Her career began in 2003 after being discovered by a freelance photographer on Bournemouth beach, and has appeared in such publications as the Daily Star tabloid newspaper and magazines FHM, Loaded, and Nuts.

If you like my work and you feel like it, you can invite me to a Ko-fi!

[This is a request]

Key word: Quinne SG

Using analogMadness v40 for all the photos.

DMP++ 2M Karras

20-30 steps

Use as you wish. Asking permission for AI generated material is stupid. Since everything is taken from someone else. Long live pirates! F them greedy..

Key word: Bae SG

Using analogMadness v40 for all the photos.

DMP++ 2M Karras

20-30 steps

Use as you wish. Asking permission for AI generated material is stupid. Since everything is taken from someone else. Long live pirates! F them greedy..

中国画风的助手，可以画出一些特定的中国画风格

需要lora配合使用: https://civitai.com/models/69771/chinese-painting-style

需要TI配合使用:https://civitai.com/models/68118/clothing

English

Assistant of Chinese painting style, can draw some specific Chinese painting styles

It needs to be used with lora: https://civitai.com/models/69771/chinese-painting-style

It needs to be used with TI: https://civitai.com/models/68118/clothing

建筑样式的textual inversion版本，需要大模型有才能召唤出来

介绍：

为什么：多个lora配合非常容易出错，但是Textual Inversion可以很好一起工作

例如：changting, heibai, (1girl:1.21), solo

主要模型：

changting: Chinese bridge (中式拱桥)

gongqiao: Lake and Pavilion(湖和亭)

guzhen: Chinese ancient town(古镇)

zhongshinei: room(室内)

其他：

安装：放到 stable-diffusion-webui\embeddings

建议配合: hanfu https://civitai.com/models/15365/hanfu

English：

The textual inversion version of the architectural style requires a large model to be summoned

Why: Multiple lora coordination is very error-prone, but Textual Inversion can work well together

For example: changting, heibai, (1girl:1.21), solo

changting: Chinese bridge

gongqiao: Lake and Pavilion

guzhen: Chinese ancient town

zhongshinei: room

Suggested cooperation: hanfu https://civitai.com/models/15365/hanfu

Installation: Put it in stable-diffusion-webui\embeddings

一些衣服样式

更容易使用的衣服召唤方法

这个版本会持续更新

为了不太干扰出图的泛化，向量为1，如果需要增强权重可以这样写 (heibai:1.30), 1girl, lace,

heibai: 黑白搭配

longpao: 刺绣样式

neiyi: 内衣样式

nvdi: 女帝

English

some clothing styles

Easier to use clothing summoning method

This version will continue to be updated

In order not to interfere with the generalization of the graph, the vector is 1. If you need to enhance the weight, you can write it like this (heibai:1.30), 1girl, lace,

heibai: black and white collocation

longpao: embroidery pattern

neiyi: underwear style

nvdi: empress

Key word: Enrapture SG

Using analogMadness v40 for all the photos.

DMP++ 2M Karras

20-30 steps

Use as you wish. Asking permission for AI generated material is stupid. Since everything is taken from someone else. Long live pirates! F them greedy..

[This is a request]

Key word: Casanova SG

Using analogMadness v40 for all the photos.

DMP++ 2M Karras

20-30 steps

Use as you wish. Asking permission for AI generated material is stupid. Since everything is taken from someone else. Long live pirates! F them greedy..

[This is a request]

Key word: Mewes SG

Using analogMadness v40 for all the photos.

DMP++ 2M Karras

20-30 steps

Use as you wish. Asking permission for AI generated material is stupid. Since everything is taken from someone else. Long live pirates! F them greedy..

A request of @tibbydapug252. (Sorry it's taken so long!)

Karen McDougal is a model and actress. She was named Playmate of the Year by the Playboy magazine in 1998.

This is a 1020-step TI trained on a dataset of 18 images with my usual settings.

Appreciate my work? My TIs are free, but you can always buy me a coffee. :)

Curious about my work process? I have summarized it here.

How to create a good prompt using my TIs

You're obviously free to experiment, but bear in mind that my TIs are trained with a more or less fixed phrasing, that normally starts with:

"photo of EMBEDDING_NAME, a woman"

So I recommend always starting your prompt like that and then building the rest of the prompt from there. For instance, "photo of beautiful (kmcd0ugal:0.99), a woman as a movie star, hair upsweep updo, sweater off-shoulders, trousers, at a movie premiere gala, dark moody ambience (masterpiece:1.2) (photorealistic:1.2) (bokeh) (best quality) (detailed skin:1.2) (intricate details) (nighttime) (8k) (HDR) (cinematic lighting) (sharp focus), (looking at the camera:1.1), (closeup portrait:1.1)"

Textual Inversion embedding I did of actress Cameron Diaz. Used about 32 reference photos of her from photo shoots, walking, etc.

I really liked how this one came out. I feel like it shows the style of her.

This model is inspired by Tiffen's Diffusion Filters and Prism Lens FX, which basically apply a layer of haze to saturated highlights in photographs. The effect is also known as Haze Light. ✨

I tried it with the checkpoints that appear in the sample images and the one I like best is ChillOutMix. 😄

It works best with white clothing and oversaturated scenes. 📸

Textual Inversion embedding I did of actress Brittany Murphy. Used about 34 reference photos of her from photo shoots, walking, etc.

I think its an ok, but seems to focus a lot more on the lips. Might do a re training with different images to get a closer view but it seems ok for now.

Textual Inversion embedding I did of singer Christina Aguilera. Used about 36 reference photos of her from photo shoots, walking, etc.

A commission of @Zorglub.

Frankie Adams is a Samoan actress, mainly known for her role as Bobbie Draper in the sci-fi TV show The Expanse.

Update: Step 1320 seems slightly better compared to the previous version, so here it is.

The TI has been trained on a dataset of 18 images with my usual settings.

Appreciate my work? My TIs are free, but you can always buy me a coffee. :)

Curious about my work process? I have summarized it here.

How to create a good prompt using my TIs

You're obviously free to experiment, but bear in mind that my TIs are trained with a more or less fixed phrasing, that normally starts with:

"photo of EMBEDDING_NAME, a woman"

So I recommend always starting your prompt like that and then building the rest of the prompt from there. For instance, "photo of (fr4nki34d4ms:0.99), a woman, RAW, close portrait photo, long brown coat, turtleneck, long haircut, slim body, (high detailed skin:1.2), 8k uhd, dslr, soft lighting, high quality, film grain, Fujifilm XT3 sharp focus, f 5.6"