Hi, any plan on stablediffusion XL version?

[lucasjinreal]

It would be must more better if there have a XL version distillation

[harishprabhala]

Yes, we are actively working on it.

[lucasjinreal]

@harishprabhala Awesome!

I have tested on the small version, the generated result looks a little bit blurry, can not have a extremly clear effect compare with original, even compare with 45+ steps.

Does this issue can be resolve with some larger params? (btw, the speed is quite fast)

[veryvanya]

what kind of speed boost are we talking? Does this utilise TensorRT or is that something that could also be added?

[Gothos]

@harishprabhala Awesome!

I have tested on the small version, the generated result looks a little bit blurry, can not have a extremly clear effect compare with original, even compare with 45+ steps.

Does this issue can be resolve with some larger params? (btw, the speed is quite fast)

Please try using negative prompts, and varying the CFG scale. That said, the model is limited by the parameter size. Perhaps some further training might help, but there are no plans for that as of now.

[lsjiiia]

@harishprabhala hi, I find that the SSD-1B model has been put on the shelves. I want to know when SDXL distilled models and code could be released?

[Gothos]

We're chalking up a technical paper for this, we plan on releasing any distillation code after. SSD 1-B is our rendition of a SDXL distilled model.

[lsjiiia]

We're chalking up a technical paper for this, we plan on releasing any distillation code after. SSD 1-B is our rendition of a SDXL distilled model.

Look forward it! Thank you!