search

horseee / DeepCache

[CVPR 2024] DeepCache: Accelerating Diffusion Models for Free
https://horseee.github.io/Diffusion_DeepCache/
Apache License 2.0
767 stars 36 forks source link

Could the author show how to control the throughput? #36

Open jinchaozhu opened 5 months ago

jinchaozhu commented 5 months ago

As show in Tab.3 , PLMS-25 steps and ours in COCO2017 is close. May be to adjust the N value to control Throughput