Logo
Explore Help
Sign In
tqcq/llama.cpp
0
0
Fork 0
You've already forked llama.cpp
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-18 16:47:42 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
3e92f4ecbe332fde1f40d4d47dff107a6544e15b
llama.cpp/pocs/CMakeLists.txt

15 lines
216 B
CMake
Raw Normal View History

Adding a simple program to measure speed of dot products (#1041) On my Mac, the direct Q4_1 product is marginally slower (~69 vs ~55 us for Q4_0). The SIMD-ified ggml version is now almost 2X slower (~121 us). On a Ryzen 7950X CPU, the direct product for Q4_1 quantization is faster than the AVX2 implementation (~60 vs ~62 us). --------- Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
2023-04-18 21:00:14 +02:00
# dependencies
find_package(Threads REQUIRED)
# third-party
include_directories(${CMAKE_CURRENT_SOURCE_DIR})
if (EMSCRIPTEN)
else()
ggml : add support for dynamic loading of backends (#10469) * ggml : add support for dynamic loading of backends --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2024-11-25 15:13:39 +01:00
if (NOT GGML_BACKEND_DL)
add_subdirectory(vdot)
endif()
Adding a simple program to measure speed of dot products (#1041) On my Mac, the direct Q4_1 product is marginally slower (~69 vs ~55 us for Q4_0). The SIMD-ified ggml version is now almost 2X slower (~121 us). On a Ryzen 7950X CPU, the direct product for Q4_1 quantization is faster than the AVX2 implementation (~60 vs ~62 us). --------- Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
2023-04-18 21:00:14 +02:00
endif()
Reference in New Issue Copy Permalink
Powered by Gitea Version: 1.24.2 Page: 8280ms Template: 84ms
English
Bahasa Indonesia Deutsch English Español Français Gaeilge Italiano Latviešu Magyar nyelv Nederlands Polski Português de Portugal Português do Brasil Suomi Svenska Türkçe Čeština Ελληνικά Български Русский Українська فارسی മലയാളം 日本語 简体中文 繁體中文(台灣) 繁體中文(香港) 한국어
Licenses API