No ComfyUI support?

[sanebg]

It's shame for a work put into this, if there is no easy way people to try it out.

[drphero]

This repository accompanies a research paper and provides the reference implementation of the method. Its purpose is to make the research reproducible, not to ship ready-made integrations for every third-party UI ecosystem.

ComfyUI support has already been mentioned by the authors as something they plan to add, but even if it weren’t, it wouldn’t be reasonable to demand it in this entitled way. The authors have already done the community a service by releasing the code publicly.

If ComfyUI integration is important to you (it is to me as well), that’s exactly the kind of thing the open-source community can contribute. Feature requests are probably welcome, but framing the absence of a convenience integration as a “shame” isn’t constructive.

[sanebg]

It's a shame because the work you put and the authors is going to waste, getting no traction because of obstacles that could've been overcome in a first place. Giving this a ComfyUI support would've exploded in starts on Github and Reddit instead of rely on some random good dude to integrate it. I'm not saying this from an entitled POV , I'm saying this as someone who's genuinly feeling bad for all the work that will get buried like many other good works just because of little nuances like this.

[drphero]

I think it's the title of this issue that is the problem. If you can edit it, that would probably help a lot with the tone.

Their hard work was for research at their universities and initially giving all the extra time and effort it would take to have an integration with ComfyUI wouldn't make sense in this scenario. If they have the time, I do hope that they can create a ComfyUI native implementation because that community would love this and would certainly help to bring more eyes to this.