π¨ Krea 2 | Mature-X [Turbo]
Small-Tune ver.
π About the Model
Base model: Krea 2 Turbo
Precision: Full Precision BF16
Model size: 24 GB
Prompting: Natural Captions & Booru Tags
Full BF16 precision means full compatibility with LoRA. The model works best with them: styles, characters, and more.
π― YES, baked-in MATURE styles. The model has mixed-in styles.
πΌοΈ Small Tune. The dataset was expanded with a tuned set of 2,000 high-quality images. It's a small tune that barely affects overall quality.
π·οΈ Flexible prompting. Works with both Captions and Booru Tags.
π A good replacement for Illustrious or Anima models.
This is the second release, so I haven't run any big experiments with the model yet.
π NEW
1. π Re-release in BF16
The model will be reassembled in BF16 quantization and released.
2. π Regarding Uncensored
I generated [R, R+, X] content perfectly fine, but not [XXX]. Even the Base model won't let you do that.
To generate full-fledged 18+ content, there are LoRAs to unlock it, for example this one:
I won't bake them in, this is your own choice.
(Breasts, nudity, genitals β pass generation and censorship on this model.)
3. πΌοΈ Regarding preview images and the lack of prompts
I won't change my opinions, but I'll make a concession: the first 3 images will always have prompts for my resources.
(I don't like sharing them. It's my personal decision, and it will remain so.)
If you don't like this approach, or you have an obsessive mania that you're being deceived, then look for other authors and their models. I'm not going to deceive anyone here or even try to.
I don't use any LoRAs or modifications in the previews to influence the final result β just simple and direct image generation.
4. π‘οΈ Merge Authorship Protection
Merging: prohibited, if the model does not declare permission for such actions on Civit or other platforms.
(Many people don't even read the permissions and prohibitions, so I integrated a small protection.)Quantization: remains and will always be available to you.
5. π¦ Personal quantization requests
If many people ask for personal quantization into certain formats that you can't do yourself, and the request is supported, of course I will release that model as soon as possible.
βοΈ Recommended Settings
My settings for Krea 2 Turbo
CLIP Skip: None (or 1)
CFG Scale: 1 β 1.5 (max)
Sampler: Euler
Scheduler: SGM_Uniform, Beta, or Simple β best way!
Steps: 8 β 10 (max)
Hi-Res Fix: β I don't use it, but you can if you want.
ADetailer: β I don't use it, but you can if you want.
π¬ Feedback and example images are very welcome. Please leave a review and share your results!
π License
The model follows the official license from the creators, without violating its copyright.
You can review the official license here:
Description
Comments (3)
I absolutely love this model! πβ€οΈ Iβm desperately waiting for an FP8 or INT8 version so I can run it on my GPU. I canβt wait! ππ₯
FP8 was already published here earlier, but people requested BF16. You can find the FP8 version here:
https://yodayo.com/models/f858d0d8-0003-4e6a-b076-1878e9319617?index=1&modelversion=b2abec26-33c3-4c2a-8e52-c26fd26b3d0a
@ArtShumovΒ ty



















