The llama.cpp project has released version b9803, which includes a fix for OpenCL to flush profiling batches at shutdown for incomplete batches. This update provides binaries for macOS, Linux, Windows, Android, and openEuler across various hardware backends.

  • macOS Apple Silicon (arm64) and Intel (x64) builds are available, with KleidiAI disabled.
  • Linux supports Ubuntu x64 and arm64 CPU, Vulkan, ROCm 7.2, OpenVINO, and SYCL FP32/FP16.
  • Windows offers CPU, CUDA 12.4/13.3, Vulkan, OpenVINO, SYCL, HIP, and OpenCL Adreno builds.
  • Android arm64 (CPU) and iOS XCFramework binaries are included.
  • openEuler supports x86 and aarch64 with ACL Graph, while the standard openEuler pull request is disabled.

The release ensures compatibility across diverse hardware configurations and corrects profiling data handling for OpenCL users.