The llama.cpp project has published the b9822 release, providing pre-built binaries for macOS, iOS, Linux, Android, and Windows. This update includes a fix for the test-chat-template --no-common option and distributes builds across various hardware architectures and accelerators.

  • macOS Apple Silicon (arm64) and Intel (x64) binaries are available, with KleidiAI support disabled.
  • Linux builds cover Ubuntu x64 and arm64 CPU, s390x CPU, Vulkan, ROCm 7.2, OpenVINO, and SYCL FP32/FP16.
  • Windows releases include CPU, OpenCL Adreno, CUDA 12.4 and 13.3, Vulkan, OpenVINO, SYCL, and HIP variants.
  • Android arm64 (CPU) binaries and an iOS XCFramework are included in the distribution.
  • openEuler builds for x86 and aarch64 architectures are listed, with some configurations disabled.

The release enables users to run llama.cpp on a wide range of devices and GPU accelerators without compiling from source.