llama.cpp b9733 update enhances ggml-webgpu and macOS/iOS support

Core summary

One sentence to understand this update

llama.cpp b9733 has been released, adding adapter toggles for F16 on Vulkan + NVIDIA for ggml-webgpu and improving support for macOS Apple Silicon and iOS XCFramework.

Impact & opportunity

What this could mean

This update provides llama.cpp developers with broader hardware compatibility and optimized performance, particularly when deploying local LLMs on Apple Silicon and mobile devices.