Iwan Kawrakow 9354ea22f6 Try interleaving 8 iq4_xs rows ...

It is also faster on AVX2.

This is the NEON implementation. It is tiny bit faster than
4 interleaved rows (~0.5%).

So, this looks like a winner given the Zen4/AVX2 improvement
without associated NEON egression.

2025-01-25 15:17:23 +02:00

..

cmake

Merge mainline llama.cpp (#3)

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

include

CPU Flash Attention improvements (#172)

2025-01-15 18:19:22 +02:00

src

Try interleaving 8 iq4_xs rows

2025-01-25 15:17:23 +02:00

.gitignore

Merge mainline llama.cpp (#3)

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

CMakeLists.txt

Move to c++17 projectwide (#80)

2024-10-04 14:43:26 +03:00