C666D Character - Natasha Kotova Krea2

C666D NATASHA KOTOVA Krea2.

Tasha: Hi, I’m… ugh, whatever 🙄, you’ll figure me out eventually. As I’m still figuring me out too. 🤷‍♀️ One day I’m a goth menace in velvet 🧛🦇, and the next I’m a pink‑ribbon coquette princess 👸👑, and sometimes I’m in business wear, acting like I run the entire city 👩‍💼💻. I change aesthetics like I change my mood... Instantly and without a warning. 🚫

I’m obsessed with looking good. Looksmaxxing. 💇 Uhhu... Makeup? 💋 Yes. Shopping? 🛒Always. My looks are basically my full‑time job, and honestly, I’m amazing at it. I don’t even know what I actually like yet, so I just try out everything. If it’s chaotic and crazy, I’ve probably done it already. 🤪

I follow my heart ❤️✔️, instead of my brain 🧠❌, because that's not where my dreams happen, no... My brain is boring, but my heart burns with passion. Does it get me into trouble? Sure. 😅 But do I care? Never! 🙄 I kind of love the mess. It keeps life interesting. 😁

I’m playful, a little spoiled, and yeah, I can be bratty sometimes, but not a single person can call me lazy. If something needs to be done, I’ll get down and dirty to make it happen. I’ll complain the entire time, but I’ll still do a better job at it than you or anyone else. 💪

And when I want something? I get it... Always! I don’t care how. I don’t care what it takes. I’m absolutely persistent, delusional, and fun too, so the holy trinity. 🤪

So yeah, I’m cute, sometimes a little arrogant, and still figuring out who to become. But no matter who I end up being, it's lowkey gonna be iconic, obviously... ❤️❤️❤️

V2 has alot of issues fixed, and also is capable of producing photo quality.
Use strength: 0.8-1.0.

Triggers: 🔥🔥🔥

1. natasha kotova,

2. a 18 year old russian girl, cute, (pre-bimbo stage).

3. a 30 year old russian woman, mature and adult, (post-bimbo stage).

4. big lips, (younger version lips).

5. big swollen lips, (mature version lips).

Quality: 🚀✈️🦅

1. a (amateur/professional/other) photo of: ,(followed by the name).

2. filmgrain texture, (to achieve photo quality).

3. realistic, (for a rendered style).

Eyecolors: 👁️🧿👁️‍🗨️

1. blue eyes,

2. green eyes,

3. grey eyes,

4. Other colors also work.

Makeup-Lips: 💄💋💄

1. shiny "color" lipstick

2. glossy "color" lipstick

3. matte "color" lipstick

Makeup-Eyeliner: 🖌️👁️🖌️

1. black eyeliner, (outer eyelid)

2. white eyeliner, (outer eyelid)

3. black and white eyeliner, (inner + outer eyelids)

Makeup-Cheeks: 😊😚😊

1. soft natural pink blush on her cheeks,

2. soft natural red blush on her cheeks,

3. Natural freckles on her nose and cheeks, (for redhead look).

Clothing: 👗👠👜

1. velvet dress with laces in front,

2. blue gingham dress,

3. pink dress with laces and bows,

4. simple ribbed white shirt,

5. rainbow croptop with on it the printed text: "daddy's girl.",

Hairstyles: 👩‍🦰👩‍🦳👱‍♀️

1. twintails,

2. ponytail,

3. hair pulled back,

4. thin braids frame her face,

5. bangs with twintails,

6. bangs with ponytail,

Accessories: 🎀💍📿

1. earrings with a heart shaped pendant,

2. large gold hoop earrings

3. rhinestone collar on her neck,

4. black leather collar with silver spikes,

Notes: 💌💌💌

1. For the dataset I used both natural language, and some pony/illu style tokens here and there, which from my experiments helps support more diverse prompt styles. So you can get messy while prompting.

2. Character is made artificially and based of a sketch I made in clipstudio when I was a active artist. And is not based of any real person. Any likeness is purely coincidental. I have the dataset to back my claim.

3. You can use all triggers of both the younger and older version interchangably.

4. All images represent a adult individual. And it's not allowed to upload the character looking as a minor.

5. I captioned the dataset manually, and focused on using different structures for sentences to describe the character.

6. If you want to get the character with the same face, use the exact make-up triggers I listed, most LLM's usually change the triggers and expand them, which can make them fail.

Updates:

1. Name trigger was not necessary with any model I tried. Using the face and make-up descriptions does the heavy lifting. But when used with other lora it helps.

2. Fixed problems with angles such as: 'from a low angle'.

3. Fixed the skin texture, looks alot less plastic.

4. V2 has a much more precise prompt response, so in V1 you could get things you didn't prompt for, this has decreased with this version, and might even need more descriptive pompting. Try it first.

5. Got rid of a bug that made a star appear.

6. Fixed issues with proportions and anatomical errors.

7. Added better support night shots. But this still underperformed in my tests (using turbo base).

8. Supports more skincolors.

9. supports more poses. But might need more deliberate instructions for the pose.

10. Increased stability and consistency in combination with other lora.

11. Increased consistency of her bodytype.

12. Thanks to the people that supported the model in early access I decided to fix her. However training Natasha already costed me 4800 buzz, which is 6 times what the model earned me. So I'm not training her again anytime soon. :'D

Help me power up: ⚡👍⚡

Hit the like button on the model page, and on my images if you enjoy what I create! And posting a image to the modelpage grants me ⚡50 blue buzz for each user that uploads to a model. Also saving my images to your gallery helps me out a little bit! So I would appreciate it alot if you could spare a few seconds! ⏰ I'm always short on buzz, and every bit helps so I can keep going! Donations are welcome too! ❤️