Apple Showcases New AI-Powered Apps Built With Foundation Models Framework

Apple is highlighting how developers are using its new Foundation Models framework to bring new AI-powered features to their apps, following the public release of iOS 26 and iPadOS 26 and macOS Tahoe 26 this month. The framework gives developers direct access to the on-device large language model at the core of Apple Intelligence.

The on-device processing allows for new features that are available offline, protect user privacy, and come at no AI inference cost to developers. The new capabilities are already appearing in apps across health, education, and productivity.

"We're excited to see developers around the world already bringing privacy-protected intelligence features into their apps," said Susan Prescott, Apple's vice president of Worldwide Developer Relations. "The in-app experiences they're creating are expansive and creative, showing just how much opportunity the Foundation Models framework opens up."


Fitness app SmartGym, for example, is using the framework to generate structured workout routines from natural language descriptions and provide personalized coaching feedback. The journaling app Stoic can now generate hyper-personal prompts based on a user's recent entries, with all processing done on-device to maintain privacy.

In the education space, the biology app CellWalk now generates conversational explanations of scientific terms tailored to a user's knowledge level. The productivity app Stuff can turn spoken or handwritten notes into organized tasks, automatically populating details like dates and tags. Video editor VLLO is also using the framework, alongside Apple's Vision framework, to intelligently analyze video clips and suggest suitable background music and stickers.

The Foundation Models framework is integrated with Swift and gives developers access to a 3 billion parameter on-device model. It is available on any Apple Intelligence-compatible device running iOS 26, iPadOS 26, or macOS Tahoe 26.