The llama.cpp project has released version b9825, which includes a fix for the Vulkan step operator when handling zero inputs. This update provides pre-built binaries for macOS, Linux, Windows, Android, and openEuler across various hardware backends.

  • macOS Apple Silicon (arm64) and Intel (x64) builds are available, with KleidiAI support disabled.
  • Linux binaries cover CPU (x64, arm64, s390x), 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 backends.
  • Android arm64 (CPU) and iOS XCFramework builds are provided for mobile deployment.
  • openEuler support includes x86 and aarch64 variants using ACL Graph, though the main pull request is marked disabled.

This release ensures broader compatibility across different operating systems and hardware accelerators while correcting specific Vulkan execution errors.