vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llava : MobileVLM support (#4954)

Browse source 
* MobileVLM native implementation

* delete depthwise_conv_2d and permute_cpy relative code, replace the two by the existed functions, and opt ldp definition, support LLAMA_PERF option for CMake

* move android script to example/llava directory

* Fix the editor config checks

---------

Co-authored-by: Chenxiaotao03 <chenxiaotao03@meituan.com>
This commit is contained in:
XiaotaoChen 2024-01-22 21:09:35 +08:00 • committed by GitHub
parent b2d80e105a
commit 3ce7e8f8e7
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 737 additions and 24 deletions
Download patch file Download diff file Expand all files Collapse all files

7
CMakeLists.txt
View file

@ -108,6 +108,13 @@ option(LLAMA_BUILD_TESTS "llama: build tests" ${LLAMA_STA
option(LLAMA_BUILD_EXAMPLES "llama: build examples" ${LLAMA_STANDALONE})
option(LLAMA_BUILD_SERVER "llama: build server example" ON)
# add perf arguments
option(LLAMA_PERF "llama: enable perf" OFF)
if (LLAMA_PERF)
add_definitions(-DGGML_PERF)
endif()
# Required for relocatable CMake package
include(${CMAKE_CURRENT_SOURCE_DIR}/scripts/build-info.cmake)