Downloads and Resources

Core ML framework reference

Browse the latest documentation including API reference, articles, and sample code.

View documentation

Core ML converters

Convert models from third-party training libraries into Core ML using the coremltools Python package.

Get started

Create ML framework reference

Browse the latest documentation including API reference, articles, and sample code.

View documentation

Metal plugin for TensorFlow

The TensorFlow Metal plugin has achieved its 1.0 release with a streamlined installation process, a graph optimizer, and mixed-precision support using 16-bit float and newly introduced 16-bit brain float datatypes.

Get started

PyTorch training on Mac

With extensive operator and network coverage, improved testing, and performance optimizations, the MPS backend has achieved its beta release. New features, such as profiling, custom kernels, and mixed-precision, enable you to accelerate and optimize more ML models.

Get started

Accelerated JAX training on Mac

The new JAX Metal plugin uses the OpenXLA compiler and PjRT runtime to GPU-accelerate JAX machine learning workloads on Mac platforms.

Get started

Explore more

Videos

Get an overview of the capabilities of Core ML presented at WWDC and other events.

View session videos

Forums

Ask questions and discuss development topics with Apple engineers and other developers.

View forums

Design guidelines

Learn how to optimize the design and user experience of your app for machine learning.

Human Interface Guidelines

Machine learning research at Apple

Read the latest machine learning research from Apple. Get updates on upcoming events and stay up to date on all news about machine learning.

Learn more