[enhancement]: Add support for PiD - Pixel Diffusion Decoder

[jackofallbrakes]

Is there an existing issue for this?

• I have searched the existing issues

Contact Details

No response

What should this feature add?

A recent plug-and-play diffusion decoder that replaces VAE/RAE decoders, turning latent representations directly into super-resolved pixels in a single pass: https://github.com/nv-tlabs/PiD

It would be great to have this in Invoke.

Alternatives

No response

Additional Content

No response

[MannXo]

Can I take this? I have some time to work on it this week.

[Pfannkuchensack]

Can I take this? I have some time to work on it this week.

You can sure try this. But it looks a bit of a big feature thing.

[MannXo]

After looking more closely at the scope, I'll sit this one out. thanks for the heads up!