apple/coreai-models

Model export recipes, Python primitives, and Swift runtime utilities for on-device AI

What it solves

This project provides the necessary tools to bridge the gap between training AI models in PyTorch and deploying them efficiently on Apple silicon devices. It simplifies the process of converting, optimizing, and running models on macOS and iOS using the Core AI framework.

How it works

It provides a three-part ecosystem:

  1. Export Recipes: Scripts to convert popular open-source models (like those from Hugging Face) into the .aimodel format.
  2. Python Primitives: Building blocks for developers to author custom models in PyTorch that are compatible with Apple's hardware.
  3. Swift Runtime: A package that allows developers to integrate these exported models directly into their Swift-based applications.

Additionally, it includes "Skills"—plugins for coding agents (like Claude Code or Codex) that provide expert guidance on model authoring, compression, and deployment workflows.

Who it’s for

This is for developers and machine learning engineers who want to build high-performance, on-device AI applications for Apple platforms using PyTorch models.

Highlights

  • Model Catalog: Includes ready-to-use export recipes for various open-source models.
  • Agentic Skills: Provides plugins for coding agents to help automate model authoring and compression tasks.
  • Native Integration: Includes a Swift package for seamless runtime execution on macOS and iOS.
  • Optimization Tools: Offers utilities for exploring weight compression, quantization, and palettization to ensure models run efficiently on-device.

Related

  • Project
    john-rocky/CoreML-Models

    A comprehensive collection of pre-converted Core ML models and sample apps for implementing on-device AI on iOS and macOS.

  • Project
    apple/coremltools

    A Python package for converting machine learning models from frameworks like PyTorch and TensorFlow into the Core ML format for on-device deployment on Apple devices.

  • Dispatch
    Apple Core AI Framework: On-Device AI Integration and Optimization

    Apple's Core AI framework provides a new way to convert PyTorch models for high-performance execution across CPU, GPU, and the Apple Neural Engine (ANE), shifting the focus toward local, on-device AI.

  • Project
    john-rocky/coreai-model-zoo

    CoreAI‑Model‑Zoo provides ready‑to‑download `.aimodel` bundles (LLMs, multimodal models) for Apple’s Core AI runtime (iOS 27/macOS 27). It includes conversion recipes, verification scripts, Swift integration via CoreAIKit, and community‑contributed ports, letting developers run sophisticated models locally on iPhone, iPad or Mac.

  • Project
    rudrankriyam/Foundation-Models-Framework-Lab

    A native iOS and macOS workbench for testing and shipping with Apple's Foundation Models framework, featuring a prompt playground, tool integration, and adapter comparison.