Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Request] GPEN Support? #2

Open
Kinglord opened this issue Oct 9, 2023 · 1 comment
Open

[Request] GPEN Support? #2

Kinglord opened this issue Oct 9, 2023 · 1 comment

Comments

@Kinglord
Copy link

Kinglord commented Oct 9, 2023

Hey there, I know this is an odd request since this is codeformer focused, but I thought I'd at least ask to see if you could add support for GPEN as well since they released some higher resolution models this year and I don't see much support for them in SD products.

https://github.com/yangxy/GPEN

I saw some reviews and the 512 model seemed to perform pretty well, so I'm curious what the 1k and 2k models can do. I actually tried to load them in but there was a Comfy error about an unsupported model, so maybe this is more of a ComfyUI request?

If you're not interested in this then that's totally cool, I just didn't find any other Facerestore focused Comfy project that wasn't targeted at faceswapping so thought I would ask. Feel free to just close this out if it's not something you care about!

@mav-rik
Copy link
Owner

mav-rik commented Oct 9, 2023

@Kinglord I would be happy to add support for GPEN, but I am not the author of this custom node, I only managed to add a codeformer fidelity parameter support to it (the original node was found on civitai as .zip archive with no refs to any repo). I am quite far from python and AI, I doubt that I am capable of adding GPEN support, at least at my current situation. If you are capable of adding GPEN support, pull request is more than welcome 🤗

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants