(1) We released the 50 diffusion steps model (instead of 1000 steps) which runs 20X faster with comparable results. (2) Calling CLIP just once and caching the result runs 2X faster for all models.
Note: Requires CPU inference (select CPU in LCMLoader Node). (facing error that i dont know how to fix when using GPU) Add '+' for more effect and '-' for less effect ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback