search

horseee / DeepCache

[CVPR 2024] DeepCache: Accelerating Diffusion Models for Free
https://horseee.github.io/Diffusion_DeepCache/
Apache License 2.0
687 stars 32 forks source link

Any plans to support SDXL? #1

Closed kilimchoi closed 7 months ago

kilimchoi commented 7 months ago

Do you guys have any plans to support XL-based models?

horseee commented 7 months ago

Hi @kilimchoi. Yes, we have plans and are currently implementing the code for Stable Diffusion XL. Once we finish and test our code, we will release the code😃

Fannovel16 commented 7 months ago

Does this method work on SDXL Turbo?

VainF commented 7 months ago

Do you guys have any plans to support XL-based models?

Hi @kilimchoi, we have released the code for SDXL. The refiner is currently not included in the demo but we are working on that. Thanks for the issue.

VainF commented 7 months ago

Does this method work on SDXL Turbo?

Hello @Fannovel16, DeepCache can only accelerate multiple-step models. So, it is not applicable to SDXL Turbo if the inference step is only 1.