Merge branch 'main' into fix/server_llama_call_thread_starvation

Author: gjpower

CHANGELOG.md

@@ -7,6 +7,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.3.2]
+
+- feat: Update llama.cpp to ggerganov/llama.cpp@74d73dc85cc2057446bf63cc37ff649ae7cebd80
+
 ## [0.3.1]
 
 - feat: Update llama.cpp to ggerganov/llama.cpp@c919d5db39c8a7fcb64737f008e4b105ee0acd20

CMakeLists.txt

@@ -6,6 +6,10 @@ option(LLAMA_BUILD "Build llama.cpp shared library and install alongside python
 option(LLAVA_BUILD "Build llava shared library and install alongside python package" ON)
 
 function(llama_cpp_python_install_target target)
+    if(NOT TARGET ${target})
+        return()
+    endif()
+
     install(
         TARGETS ${target}
         LIBRARY DESTINATION ${CMAKE_CURRENT_SOURCE_DIR}/llama_cpp/lib
@@ -75,7 +79,22 @@ if (LLAMA_BUILD)
     add_subdirectory(vendor/llama.cpp)
     llama_cpp_python_install_target(llama)
     llama_cpp_python_install_target(ggml)
-    
+
+    llama_cpp_python_install_target(ggml-base)
+
+    llama_cpp_python_install_target(ggml-amx)
+    llama_cpp_python_install_target(ggml-blas)
+    llama_cpp_python_install_target(ggml-can)
+    llama_cpp_python_install_target(ggml-cpu)
+    llama_cpp_python_install_target(ggml-cuda)
+    llama_cpp_python_install_target(ggml-hip)
+    llama_cpp_python_install_target(ggml-kompute)
+    llama_cpp_python_install_target(ggml-metal)
+    llama_cpp_python_install_target(ggml-musa)
+    llama_cpp_python_install_target(ggml-rpc)
+    llama_cpp_python_install_target(ggml-sycl)
+    llama_cpp_python_install_target(ggml-vulkan)
+
     # Workaround for Windows + CUDA https://github.com/abetlen/llama-cpp-python/issues/563
     if (WIN32)
         install(

llama_cpp/__init__.py

@@ -1,4 +1,4 @@
 from .llama_cpp import *
 from .llama import *
 
-__version__ = "0.3.1"
+__version__ = "0.3.2"