Core ML

Core ML is optimized for on-device performance of a broad variety of model types by leveraging Apple silicon and minimizing memory footprint and power consumption.

What’s new

Updates to Core ML will help you optimize and run advanced generative machine learning and AI models on device faster and more efficiently. Core ML Tools offers more granular and composable weight compression techniques to help you bring your large language models and diffusion models to Apple silicon. Models can now hold multiple functions and efficiently manage state, enabling more flexible and efficient execution of large language models and adapters. The Core ML framework also adds a new MLTensor type which provides an efficient, simple and familiar API for expressing operations on multi-dimensional arrays. And Core ML performance reports in Xcode have been updated to provide more insight into support and estimated cost of each operation in your model.

Watch the latest videos

Experience more with Core ML

Run models fully on-device

Core ML models run strictly on the user’s device and remove any need for a network connection, keeping your app responsive and your users’ data private.

Run advanced machine learning and AI models

Core ML supports generative AI models with advanced model compression support, stateful models and efficient execution of transformer model operations.

Convert models to Core ML

Models from libraries like TensorFlow or PyTorch can be converted to Core ML using Core ML Tools more easily than ever before.

Mac Studio and Apple Studio Display with Xcode windows open.

Xcode integration

Core ML is tightly integrated with Xcode. Explore your model’s behavior and performance before writing a single line of code. Easily integrate models in your app using automatically generated Swift and Objective‑C interfaces. Profile your app’s Core ML‑powered features using the Core ML and Neural Engine instruments.

Performance reports

Generate model performance reports measured on connected devices without having to write any code. Review a summary of load and prediction times along with a breakdown of compute unit usage of each operation. Save, load and compare performance reports across different devices and models optimizations.

Profile with Instruments

Profile your app to view Core ML API calls and associated models using the Core ML instrument. Find out where and when Core ML dispatches work to the hardware and gain more visibility with the Metal and Neural Engine instruments.

Live preview

Preview your model’s behavior on sample data files or live from your device’s camera and microphone, all right in Xcode.

Encrypt models

Xcode supports model encryption, enabling additional security for your machine learning models.

Powerful Apple silicon

Core ML is designed to seamlessly take advantage of powerful hardware technology including CPU, GPU, and Neural Engine, in the most efficient way in order to maximize performance while minimizing memory and power consumption.

Get started with Core ML

Create ML

Build and train Core ML models right on your Mac with no code.

Learn more

Core ML Tools

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

Learn more

Models

Get started with models from the research community that have been converted to Core ML.

Browse models