What’s New in Core ML 3

Integrate a broad variety of machine learning model types into your app. Core ML 3 enables advanced neural networks with support for over 100 layer types, and seamlessly takes advantage of the CPU, GPU, and Neural Engine to provide maximum performance and efficiency. You can run machine learning models directly on the device.

On-device Training

Core ML models are bundled into apps to help drive intelligent features like search or object recognition in photos. Now these models can be updated with user data on-device, helping models stay relevant to user behavior without compromising privacy.

Support for Advanced Neural Networks

Today's most sophisticated machine learning models can now run on-device with Core ML 3. Use and run the latest models, such as cutting-edge neural networks designed to understand images, video, sound, and other rich media.

Support for 100+ model layer types


Easily build computer vision machine learning features into your app. Take advantage of face detection, tracking, and capture quality as well as text recognition, image saliency and classification, and image similarity identification. Other features include improved landmark detection, rectangle detection, barcode detection, object tracking, and image registration. Use the new Document Camera API to detect and capture documents using the camera.

Vision Framework Reference

Natural Language

Analyze natural language text and deduce its language-specific metadata for a deep understanding. You can use this framework with Create ML to train and deploy custom NLP models. Features include transfer learning for Create ML text models, word embeddings, sentiment classification, and a text catalog. Available for English, French, German, Italian, Simplified Chinese, and Spanish.

Natural Language Framework Reference


Take advantage of on-device speech recognition for 10 languages as well as speech saliency features such as pronunciation information, streaming confidence, utterance detection, and acoustic features. Your apps can access phonetic recognition information to disambiguate homonyms. These features are now also available for Mac apps.

Speech Framework Reference