Hey everyone, I've been messing around with Krea2 and the community Identity Edit LoRA so this is my first workflow that I'm sharing publicly. I'm still pretty new to ComfyUI, so it’s probably not the most optimized or the "best" way to do this. But it worked well for me, so I figured I'd upload it in case it helps someone else.
Basically, this uses a manual inpainting approach. You have to load your image, open the mask editor, and draw the mask yourself over the exact area you want to change. The workflow then forces Krea 2 to only generate inside that mask, leaving the rest of the original image completely untouched at the pixel level.
I threw in a few extra things to make it easier to use:
A node to automatically detect the size of your input image, so you don't have to manually type in the width and height.
A mask expanding node so you can feather the edges a bit. If the model keeps making the object bigger when you just want to change its color, I recommend lowering the "expand" number or even using a negative number.
rgthree's Power Lora Loader so you can stack the Identity Edit LoRA with more LoRAs.
An image comparer node so you can slide back and forth between the original and the edit.
A quick tip: when prompting, using words like "recolor" instead of "change" helps prevent the model from redrawing the shape and making it larger than it should be.
Feel free to tweak it, break it, or let me know what I could have done better. Still learning :p
Description
FAQ
Comments (11)
Thanks for sharing your WF, will give it a try.
sure, if it works for you or find something to add/fix, leave a comment pls
@nvclei Works great! Clean and simple. I guess next would be a 2 image version?
@yajukun i’m currently working on v2 replacing native inpainting with LanInpainting (for inpainting only) + TrixLoader nodes (smart masking). I’ll totally consider adding 2 image editing. Thanks again for your comment and feedback
with model krea2 turbo int8 convrot it gives error while with krea2 turbo fp8 scaled the inpainted area looks red.
I don't have any errors with int8 convrot/fp8, did you check if your ComfyUI is updated? Those checkpoints works fine in any other workflow?
@nvclei
"Some nodes are missing required inputs
ValueError: Unknown quantization format for layer blocks.0.attn.wq"
tried a different model and it worked only with the identity v1.1 lora.
face expression changed but genitals did not, used hairy lora and bypass lora. it barely modifies it or it removes some of the pubic hair
@drfaker911219 I think it is an update issue, nodes update or comfyui, please check that
@drfaker911219 and try this Lora, instead of the bypass one.
https://huggingface.co/Beinsezii/Krea-2-Turbo-Projector-Scale-LoRA-Diffusers
I've only just updated everything, and the int8 convrot model still throws a "Key error: 'int8_rowwise'. With the fp8 model, all I get is a reddish translucent blur in the shape of my mask (that does unwanted changes to the image underneath).
works poorly unfortunately
