Swift-Llama

Run LLMs locally on Apple devices. Native Swift wrapper around llama.cpp with Metal GPU acceleration, streaming tokens, and tool call parsing.

Swift-Llama is a native Swift library for running large language models locally on Apple devices. It wraps llama.cpp with a clean Swift API, supporting Metal GPU acceleration, streaming token generation, tool call parsing, and ChatML templates. Built with Swift 6.0 concurrency (async/await actors) and distributed as a Swift Package.