llama.cpp has released version b9694, including binaries for macOS, Linux, Android, Windows, and openEuler. The release supports various architectures and acceleration options such as CUDA, Vulkan, OpenVINO, SYCL, and ROCm. A fix for the Windows x64 OpenVINO release link was also implemented.
llama.cpp releases version b9694 with new binaries for multiple platforms
from English