llama.cpp Release b9699 Adds SYCL Support and Multiple Platform Binaries

llama.cpp version b9699 introduces support for MUL_MAT and OUT_PROD operations with Q1_0 precision via PR #24721. The release includes precompiled binaries for macOS, Linux, Android, Windows, and openEuler across multiple architectures and acceleration frameworks, including SYCL (FP32 and FP16), Vulkan, CUDA, ROCm, and OpenVINO.