FaceSwap does not look like my model

[shortskirtlover]

I am having issues with getting my generation to look like my FaceSwap, I have tried every setting I can think of and she just doesn't come out the way I would like. If any of you could give me suggestions I would appreciate it. I am using the same seed, resolution, no negative prompt, same styles and models, different settings in FaceSwap. I have also tried different checkpoints and LoRA as well. I don't know if I am overcomplicating things or what.

CONTROL IMAGE:

IMAGE GUIDANCE SCALE = 5 / SHARPNESS = 2
FACESWAP = STOP 1 / WEIGHT .75
IMAGE PROMPT = STOP .5 / WEIGHT .5

IMAGE GUIDANCE SCALE = 5 / SHARPNESS = 2
FACESWAP = STOP 1 / WEIGHT .75
PYRACANNY = STOP .8 / WEIGHT .5

TOOK OUT PROMPT
IMAGE GUIDANCE SCALE = 5 / SHARPNESS = 2
FACESWAP = STOP 1 / WEIGHT .75
IMAGE PROMPT = STOP .8 / WEIGHT 1.5

RAISED SHARPNESS
IMAGE GUIDANCE SCALE = 5 / SHARPNESS = 10
FACESWAP = STOP 1 / WEIGHT .75
PYRACANNY = STOP .5 / WEIGHT 1s

[ILikeToasters]

Increase weight for FaceSwap to 1 and try. Not sure if you have tried that but didn't see it in the ones you showed. I find best luck is above .9 weight.

[lgwjames]

im using a master photo of model neck up , stop at 9 and weight 8-855 and all ok

[infinity0]

Do not try to mix content generation and faceswap in one Generate step. Do your generation first, then apply the output as an input to the method described in #2644, i.e. Upscale + FaceSwap Stop At = 1.0.