Great work so far. This has been a lot of fun to play with.

I am down to try it out

Good idea, afaik many ppl w/ low vram GPUs are struggling w/ using CN w/ highres fix so if it works it'll be great.

Great which 3 lines did you change? I would like to try it.

looks awesome , keep it up

Is this speedup competitive with T2I adapter?

that's good news for me, 期待早点体验，目前gpu vram总是不够，计算也很慢

Wouldn't this let you use a much larger model

Yes, it's really necessary. I have a 4GB GPU and I would benefit of this. Thank you for you hard work.

Good idea, I need it very much,

Could we use the released model to achieve this or should we retrain a new model to achieve this?
I tried to tweak the sampling steps in the way you described and used the released model, and it didn't work.😢

we need it =))

how to deal with the timesteps? Does it use T at the first step, and the T result for all following steps?

Update: this experiment fails to train ControlNets as good as the standard implementation. We have given up this feature. The input from each diffusion steps is necessary for robustness, and necessary for special models like Shuffle and IP2P.

can we use NAS (neural architecture search) for even more GPU efficiency ?

finding new neural architecture which is even much faster ?

does anyone have Nvidia dgx workstation to run NAS , and find a better and even more gpu efficient and fast algorithm ?