Krea2
V1 was a little unstable at high values regarding ethnicity.
That problem is fixed on V2.
V2:
Recommended range: -3.0 to 5.0
(try -10.0 or 10.0 to get a fun effect)
V1:
Recommended stable range: -1.5 to 1.5
Recommended unstable range: -4.0 to 4.0
ZIT
V2 is a huge improvement.
V2 is as close to ethnicity stable as i think is possible. As long as you stay within the range, this should not change a prompted ethnicity to something else. But it may drift within the ethnicity, like east to west.
Samples represents the LoRAs effect very well.
Only in the negative there is a tiny bit asian-bias when there is other asians in the image, as my sample with the 3 girls show.
Stable range: -3 and 3
But i will leave the range available wider just in case.
All my work is 100% free. The only thing that really matters to me is people leaving a review, good or bad. I will NEVER remove or block anyone for leaving a bad review, so don't be afraid to give me criticism.
I'm stating this because when a creator blocks someone for leaving a bad review, that person can no longer see any of the creator's models, essentially muting their critics. This is something I'm highly critical of, and I believe it's a fundamental quality problem on this site.
Description
FAQ
Comments (10)
Hello, thank you for providing so many high-quality Slider models. Could you please create a Cameltoe Slider model? I think it would be very interesting.
If you really want to make this model, it's advisable not to affect the facial. thanks！
Incredible slider models. I am convinced you are the best model creator for ZIT. Not sure if this is a slider but perhaps could you try create a LORA for "Face diversity" or "Face difference"? Ideally this is to create different faces for multiple people in an image
I love ideas like this, that seemingly is a simple idea. But when i start to think about how i could do it, i have no idea where to even begin. I will try something right away, and we will see if its even possible with a slider.
You have an amazing collection of sliders. Too bad you're not touching F2K.
I have tried, but i only have 12GB VRAM myself. It is no problem with 12GB when generating single images because of the dynamic loading in comfyui, but when i make the grids i use to test LoRAs etc its problematic. As well as needing to rent more expensive GPUs to train the models.
And when i dont use Klein myself, i dont really see the point of it. Sorry
can we get this for krea?
I can try. I got the setting for this LoRA in my archive since the results was pretty good. So i will try the same setup for krea2 and see if it works.
@Loraholic thank you!
@Loraholic You really are one of the best at this. Thank you for sharing 🙌