The llama.cpp project has released version b9959, which includes a synchronization with the ggml library. This update provides pre-built binaries across multiple platforms and hardware configurations.

  • macOS Apple Silicon (arm64) and Intel (x64) builds are available, though KleidiAI support is disabled.
  • Linux binaries cover Ubuntu x64, arm64, and s390x CPU variants, along with Vulkan, ROCm 7.2, OpenVINO, and SYCL FP32/FP16 options.
  • Windows releases include CPU, Vulkan, OpenCL Adreno, CUDA 12.4/13.3, OpenVINO, SYCL, and HIP builds.
  • Android arm64 (CPU) binaries are provided for mobile deployment.
  • openEuler support includes x86 and aarch64 variants with ACL Graph, while some openEuler configurations remain disabled.
  • A standalone UI package is also included in the release.