llama.cpp
Dequant improvements rebase
#8255
Merged

Dequant improvements rebase #8255

AidanBeltonS
Remove double lines
730819cf
Single load for half2
80d3b39b
Store scales in local mem
6c7c9379
Vec load quantized values
504a47ab
JohannesGaessler JohannesGaessler added SYCL
OuadiElfarouki
OuadiElfarouki approved these changes on 2024-07-02
joeatodd
airMeng
airMeng approved these changes on 2024-07-02
luoyu-intel
airMeng airMeng merged fadde671 into master 1 year ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone