LightCompress icon indicating copy to clipboard operation
LightCompress copied to clipboard
Published 1 month ago • verified publisher icon ModelTC

will you support two stage quantization of weight or activation as Qserve?

Open geqian-9192 opened this issue 11 months ago • 1 comments

Hi, thanky you for your good job! Since fine-grained quantization has a significant impact on the results, is it possible to support an algorithm similar to the two-stage quantization of weights in Qserve?

geqian-9192 avatar Feb 14 '25 03:02 geqian-9192

Qserve will be supported in the future.

gushiqiao avatar Feb 17 '25 11:02 gushiqiao