Type | |
Stats | 498 33 |
Reviews | (64) |
Published | Sep 26, 2024 |
Base Model | |
Training | Epochs: 30 |
Usage Tips | Strength: 0.9 |
Hash | AutoV2 763D5136B3 |
Trained on 56 images for 30 epochs.
The Lora is not quite perfect and cannot always prevent the “beloved” flux chin.
But at least it's a good start. Depending on how well the Lora is received, I will work further on it.
The aim of the whole thing was to change as little as possible of the overall look and I think I succeeded quite well.
Try out the Lora for yourself and then decide whether you like it or not.
If you use Inpainting, you can set the Lora strength to 2-3 to get rid of the flux chin.
_______________________________________________________________________________________________
Recommended Settings:
Distilled CFG Scale:
2
Sampling method and Schedule type:
I personally use DEIS/Beta.
You have to decide for yourself if other combinations deliver better results.
Steps:
30
Resolution/Aspect Ratio:
The Lora is trained on 1:1 aspect ratio.
You will have to test if other aspect ratios work.
Lora Weight:
0.5-1
But again, it all depends on your own taste.
Trigger:
No trigger
_______________________________________________________________________________________________
I would be very grateful if you could leave me your feedback here
and post pictures with your results. I would really appreciate it if you guys would leave me a small Buzz tip here so I can train more Loras.