Made this LoRA to fix the stiff, robotic face look and generic centering on Krea 2.
It helps generate subtle micro-emotions (genuine laughs, hesitation, smirks, tired looks) and more dynamic, candid framing without forcing a specific style.
This LoRA is a first release, so feel free to experiment with different weights. You'll see it can be really effective for creating more dynamic poses and natural expressions, moving away from stiff and generic results.
How to use:
Trigger word: None. Just describe the expression and pose naturally in your prompt.
Weight:
0.6-0.8(around0.8is the sweet spot).Base model: Krea 2.
Works well on photorealism, but also transfers nicely to stylized art, 3D, and anime. Feel free to share your results !
Description
Initial release! Feel free to experiment with different weights (0.6 to 1.0). It's designed to bring more dynamic framing, natural micro-emotions, and candids without rigid poses.