Kawrakow 82645c4be7 Faster IQ3_KT and IQ4_KT (#453) ...

* Somewhat faster iq3_kt (AVX2)

* Cleanup

* Slightly faster iq4_kt

* Slightly faster iq4_kt

PP is now almost 50% better than original, TG is ~20% better

* Cleanup

* Very slightly faster iq4_kt TG

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>

2025-05-24 11:48:52 +03:00

..

cmake

Merge mainline llama.cpp (#3)

2024-07-27 07:55:01 +02:00

include

Trellis quants with CPU inference (#441)

2025-05-23 09:17:52 +03:00

src

Faster IQ3_KT and IQ4_KT (#453)

2025-05-24 11:48:52 +03:00

.gitignore

Merge mainline llama.cpp (#3)

2024-07-27 07:55:01 +02:00

CMakeLists.txt

Option to enable disable the IQK CPU FA kernels (#429)

2025-05-17 11:21:58 +03:00