llama.cpp Release b12399 Introduces Vulkan Build Fixes
Release b12399 of llama.cpp introduces minor build improvements, specifically adding explicit inclusion of SPIRV-Headers for Vulkan support. It also includes workarounds for OSX CI runner configurations. These build fixes ensure smoother compilation across various Linux distributions and operating systems where header files are not located in default system paths. This helps maintain reliable cross-platform support for hardware acceleration via Vulkan. The update utilizes appropriate CMake targets to specify the SPIRV-Headers location explicitly to the compiler, addressing issues on distributions with non-standard include directories.
## BACKGROUND
llama.cpp is an open-source library that enables efficient local inference for large language models using C/C++. Vulkan is a cross-platform graphics and compute API that allows llama.cpp to leverage compatible GPUs for accelerated model execution.