Apple Intelligence and machine learning
Add intelligent features with Apple Intelligence, machine learning, and related technologies.
Apple Intelligence
Intelligent features make people’s experiences on a device more personal. Apple incorporates intelligent features into many of its frameworks and technologies, making it easy to adopt those features quickly in your own apps. When a framework doesn’t do what you need it to, you can customize its behavior by training a model with your own data — like training a new sound classification model to detect your own sound. Of course, if you have your own machine learning models, you can also incorporate them into your app in a way that works well with Apple hardware and the system’s other intelligent features.
Apple Intelligence
Apple Intelligence is the personal intelligence system that helps you communicate, work, and express yourself. It combines generative models with your personal context to deliver intelligence that is most useful and relevant to people, while designed with privacy in mind. Now you have direct access to the on-device model at the core of Apple Intelligence, so you can build experiences that are smart, private, and work even without internet connectivity.
Foundation Models framework
Run your own prompts on the on-device model at the core of Apple Intelligence to enhance your existing features with model capabilities like text extraction and summarization. When you prompt the model, provide your Swift data type that it uses to avoid making structural mistakes — allowing you to focus on your prompts and not parsing JSON schemas. The models can also call your own code to help perform additional tasks that might need real-time information or content thatʼs in a local database.
Intelligent frameworks
Many system frameworks use machine learning models to make difficult tasks easier, and make seemingly impossible tasks possible. You can use system frameworks that help with performing analysis of Vision, Speech, and Natural Language.
Custom machine learning models
Create your own machine learning models when you need more customization than what the system technologies provide. Use the Create ML app when you want a fast, user-friendly way to create and train models with your own data. If you already have a model you’ve built, you can convert it to the Core ML format with Core ML tools and deploy it into your app.