Iwan Kawrakow 8a83e1f083 cuda: re-add q8_0 -> q8_0 transpose ...

so mla = 2 can be used with CUDA graphs and q8_0 cache.

2025-08-14 19:33:58 +03:00

..

cmake

Merge mainline llama.cpp (#3)

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

include

gpt-oss: add ability to use -fmoe (only CUDA for now)

2025-08-12 09:49:18 +03:00

src

cuda: re-add q8_0 -> q8_0 transpose

2025-08-14 19:33:58 +03:00

.gitignore

Merge mainline llama.cpp (#3)

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

CMakeLists.txt

Vulkan: add cmake options to build without coopmat(2) support (#674)

2025-08-07 17:26:21 +03:00