ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-02-27 00:24:11 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
8457a26f83b2f6acd014449e91bfb60a37fcec0e
ik_llama.cpp/ggml
History
Kawrakow 8457a26f83 CUDA: faster float -> iq4_nl conversion (#73) 
* iqk_mul_mat: better iq4_nl implementation on Zen4/AVX2

PP-512 performance for LLaMA-3.1-8B goes to 162.6 t/s up
from 133.2 t/s.

* Speed up float -> iq4_nl conversion on CUDA

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
2024-10-01 12:28:29 +03:00
..
cmake
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
include
Adding SWIGLU unary op (#65)
2024-09-28 13:37:25 +03:00
src
CUDA: faster float -> iq4_nl conversion (#73)
2024-10-01 12:28:29 +03:00
.gitignore
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
CMakeLists.txt
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00