Flux2Klein_4B_diff_loras

626 downloads 0 likes on Hugging Face updated

Model card

Extracted Loras from FLUX2 Klein 4B finetunes

Model description

Loras extracted from finely tuned checkpoints. I did it primarily for myself because it’s easier to work with Lora, including mixing them in real time and setting different weights/strengths. It also saves space on the local computer. Choose a rank according to your taste and the capabilities of your hardware (the lower the rank, the less memory you need). After extraction, the weights were normalized. In most cases, ranks of 16, 32, or 64 will be sufficient. There is a correlation between rank and the acceptable Lora strength range. However, I was unable to establish the exact parameters of this correlation because it seems to be unique to each model (and I didn't have the time or inclination to thoroughly test the results). In any case, a strength range of 40-120% is where you should start experimenting. Caution! Chroma2 Kaleidoscope may require a minimum strength setting (up to 20%).

Download model

Download them in the Files & versions tab.