Iwan Kawrakow 130cdf2715 New DeepSeek FlashMLA ...

Does not work because the RoPE portion is stored at the end
in our case, while in mainline it is stored at the beginning,
and the FA kernel assumes that.

2025-05-11 09:58:03 +03:00

..

cmake

Merge mainline llama.cpp (#3)

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

include

Add copyright notices (#317)

2025-04-07 10:43:26 +02:00

src

New DeepSeek FlashMLA

2025-05-11 09:58:03 +03:00

.gitignore

Merge mainline llama.cpp (#3)

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

CMakeLists.txt

Compile time option to use bf16 for qunts without MMQ kernels (#261)

2025-03-18 07:37:10 +01:00