Sign In

Deep Cache, lcm and SDXL is so fast!

Deep Cache, lcm and SDXL is so fast!

Special thanks to https://civitai.com/user/speach1sdef178 and https://civitai.com/user/Triple_Headed_Monkey , https://civitai.com/user/Afroman4peace , https://civitai.com/user/Akatsuzi .

You can speed up ComfyUi with SDXL models quite easily, just by docking deepcache.py to the custom_nodes folder. (remember to reset the GUI).

Link do deep cache: https://gist.github.com/laksjdjf/435c512bc19636e9c9af4ee7bea9eb86, im not a author (https://arxiv.org/abs/2312.00858)

In addition, I added LCM and linked to DeepCache. As a result, the first 70 steps are generated in 15 seconds (RTX 3060 ti).

Have fun with that WF :)

Support Paweł Tomczuk on Ko-fi! ❤️. ko-fi.com/aderek514 - Ko-fi ❤️ Where creators get support from fans through donations, memberships, shop sales and more! The original 'Buy Me a Coffee' Page.

Visit my DeviantArt page: Aderek - Hobbyist, Digital Artist | DeviantArt

2

Comments