What problem does it solve?
This Skill eliminates the difficulty of building privacy-preserving AI features for iOS apps that require offline functionality and no user data to leave the device, which is not possible with cloud-dependent large language models.
Core Features & Use Cases
- On-Device Text Generation: Generate summaries, content, and responses directly on the user's device without cloud API calls, ideal for privacy-sensitive use cases like note-taking or personal assistant apps.
- Structured Data Extraction: Use the @Generable macro to extract typed, structured data from natural language input for form filling, command parsing, or data entry tasks.
- Custom Tool Calling: Implement domain-specific AI actions like recipe search or calendar lookup by defining custom tools the model can invoke automatically.
- Snapshot Streaming: Stream partially generated structured responses to update UIs in real time as the model produces output, for responsive user experiences.
- Use Case Example: A cooking app can use this Skill to extract structured dietary preferences from user natural language input, then generate personalized recipe suggestions entirely on-device without sending user data to external servers.
Quick Start
Use the foundation-models-on-device skill to add a private on-device AI feature to your iOS 26+ app that extracts structured user preferences from natural language input and returns personalized suggestions without sending data to the cloud.
