ml-explore/mlx-swift

Swift API for MLX

What it solves

MLX Swift provides a native Swift API for the MLX array framework, enabling developers to perform machine learning research and experimentation directly within the Swift language on Apple silicon.

How it works

It acts as a Swift expansion of the MLX framework, allowing users to build and train models using Swift. It supports multiple build systems including Xcode, SwiftPM, and CMake, and provides backends for Metal (on macOS/iOS) and CUDA (on Linux).

Who it’s for

Swift developers and ML researchers who want to leverage Apple silicon's performance for machine learning tasks on macOS, iOS, and Linux.

Highlights

  • Native Swift API for the MLX array framework.
  • Supports LLM and VLM implementations via the complementary mlx-swift-lm package.
  • Cross-platform support for macOS, iOS, and Linux.
  • Includes examples for training LeNet on MNIST, running chat apps with LLMs/VLMs, and generating images with Stable Diffusion.

Related

  • Project
    ml-explore/mlx-swift-lm

    A Swift package for building applications with large language models and vision language models using MLX Swift, featuring model loading, fine-tuning, and guided generation.

  • Project
    ml-explore/mlx-swift-examples

    A collection of example programs and libraries for MLX Swift, demonstrating how to build LLM, VLM, and image generation apps on iOS and macOS.

  • Project
    ml-explore/mlx

    An array framework by Apple machine learning research optimized for machine learning on Apple silicon, featuring a unified memory model and familiar NumPy/PyTorch-like APIs.

  • Project
    SharpAI/SwiftLM

    A native Swift inference server for Apple Silicon that serves MLX models via an OpenAI-compatible API, featuring SSD expert streaming and KV cache compression.

  • Project
    Blaizzy/mlx-audio-swift

    A modular Swift SDK for audio processing on Apple Silicon, providing on-device support for text-to-speech, speech-to-text, and speaker diarization using the MLX framework.