Repolode
Submit
apple

apple/coremltools

Python package for converting trained TensorFlow, PyTorch, scikit-learn, XGBoost and LibSVM models into Apple's Core ML format.

View on GitHubWebsite
Stars
5,437
Forks
868
Open issues
542
Last commit
22 days ago

What is it?

Core ML Tools (coremltools) is a Python package for converting machine learning models from third-party training libraries into Apple's Core ML format, so they can run inside iOS, macOS and other Apple platform apps. It supports TensorFlow 1.x and 2.x, PyTorch, and non-neural-network frameworks such as scikit-learn, XGBoost and LibSVM.

Beyond conversion, the package can read, write and optimize existing Core ML models. On macOS, it can also check a conversion by running predictions with Core ML. Once a model is converted, it is added to an Xcode project and called through Core ML APIs.

The package is installed from PyPI with pip and is licensed under BSD-3-Clause. Documentation includes a guide with examples and an API reference.

Key features

  • 01Converts TensorFlow, PyTorch, scikit-learn, XGBoost and LibSVM models to Core ML.
  • 02Reads, writes and optimizes existing Core ML models.
  • 03Validates conversions by running predictions with Core ML on macOS.
  • 04Installable from PyPI with pip.

Use cases

  • Bringing a PyTorch or TensorFlow model into an iOS or macOS app.
  • Deploying a scikit-learn or XGBoost model on Apple devices.
  • Inspecting or editing an existing Core ML model from Python.

Quick start

  1. 1

    Install the latest stable release from PyPI with pip

    pip install coremltools

Things to know

  • Running predictions to verify conversions requires macOS.
  • Full-featured use depends on the Core ML framework, which is Apple-platform specific.

Topics

#coreml#model-conversion#machine-learning#pytorch#tensorflow#apple#on-device-ai#coremltools#model-converter

FAQ

What is coremltools?
Python package for converting trained TensorFlow, PyTorch, scikit-learn, XGBoost and LibSVM models into Apple's Core ML format. Core ML Tools (coremltools) is a Python package for converting machine learning models from third-party training libraries into Apple's Core ML format, so they can run inside iOS, macOS and other Apple platform apps. It supports TensorFlow 1.x and 2.x, PyTorch, and non-neural-network frameworks such as scikit-learn, XGBoost and LibSVM.
Is coremltools free to use? What license does it have?
coremltools is open source under the BSD-3-Clause license. Check the license terms before using it in commercial products.
Is coremltools still maintained?
Actively maintained. The last commit to coremltools was on Sep 30, 2026.
What can I use coremltools for?
Bringing a PyTorch or TensorFlow model into an iOS or macOS app. Deploying a scikit-learn or XGBoost model on Apple devices. Inspecting or editing an existing Core ML model from Python.
How do I install coremltools?
Install the latest stable release from PyPI with pip: pip install coremltools
Where can I get coremltools?
The source code and installation instructions are on GitHub: https://github.com/apple/coremltools

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects