root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-27 03:44:35 +00:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

cmake : add helper for faster CUDA builds

Browse Source
This commit is contained in:
Georgi Gerganov 2023-10-23 20:35:19 +03:00
parent 96981f37b1
commit 8fb1be642e
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
CMakeLists.txt
View File

@ -331,6 +331,7 @@ if (LLAMA_CUBLAS)
set(CMAKE_CUDA_ARCHITECTURES "60;61;70") # needed for f16 CUDA intrinsics set(CMAKE_CUDA_ARCHITECTURES "60;61;70") # needed for f16 CUDA intrinsics
else() else()
set(CMAKE_CUDA_ARCHITECTURES "52;61;70") # lowest CUDA 12 standard + lowest for integer intrinsics set(CMAKE_CUDA_ARCHITECTURES "52;61;70") # lowest CUDA 12 standard + lowest for integer intrinsics
#set(CMAKE_CUDA_ARCHITECTURES "") # use this to compile much faster, but only F16 models work
endif() endif()
endif() endif()
message(STATUS "Using CUDA architectures: ${CMAKE_CUDA_ARCHITECTURES}") message(STATUS "Using CUDA architectures: ${CMAKE_CUDA_ARCHITECTURES}")