PiDTLN icon indicating copy to clipboard operation
PiDTLN copied to clipboard

Published 20 hours ago • verified publisher icon SaneBow

Quantitative models are slower than the original models

Open leizhu1989 opened this issue 1 year ago • 1 comments

hello, when I try in c++ project to infer the Quantized models，I find it is slower than original float32 models. why is it?

leizhu1989 avatar Dec 07 '23 01:12 leizhu1989