We’re excited to announce a new build of llama.cpp (b10159), which includes a performance-boosting addition for Apple Silicon users. This release introduces the Fast Walsh-Hadamard Transform (FWHT) kernel into the Metal backend, improving the efficiency of certain mathematical operations during model inference. While the change is low-level, it helps streamline computation on macOS and iOS devices with Metal support, contributing to faster and smoother local AI experiences.
The following pre-built packages are available for download:
Website
macOS/iOS
- macOS Apple Silicon (arm64)
- macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED]
- macOS Intel (x64)
- iOS XCFramework
Linux
- Ubuntu x64 (CPU)
- Ubuntu arm64 (CPU)
- Ubuntu s390x (CPU)
- Ubuntu x64 (Vulkan)
- Ubuntu arm64 (Vulkan)
- Ubuntu x64 (ROCm 7.2)
- Ubuntu x64 (OpenVINO)
- Ubuntu x64 (SYCL FP32)
- Ubuntu x64 (SYCL FP16)
Android
Windows
- Windows x64 (CPU)
- Windows arm64 (CPU)
- Windows arm64 (OpenCL Adreno)
- Windows x64 (CUDA 12) – CUDA 12.4 DLLs
- Windows x64 (CUDA 13) – CUDA 13.3 DLLs
- Windows x64 (Vulkan)
- Windows x64 (OpenVINO)
- Windows x64 (SYCL)
- Windows x64 (HIP)
openEuler
- [DISABLED]
- openEuler x86 (310p)
- openEuler x86 (910b, ACL Graph)
- openEuler aarch64 (310p)
- openEuler aarch64 (910b, ACL Graph)
