Apple has introduced its Foundation Models framework across iOS 26, iPadOS 26, and macOS 26, giving developers access to on-device large language models that enable private, offline AI experiences. This framework forms the core of Apple Intelligence and is already being integrated into a wide range of third-party apps highlighted by Apple.
According to the company, the framework allows developers to build features that previously required heavy cloud infrastructure. Because all processing happens on device, user privacy is maintained and inference is free of cost. Apps in health, fitness, journaling, and education are showcasing how the technology unlocks personalized and context-aware experiences.
SmartGym is using the framework to let users describe workouts in natural language and automatically generate structured routines with reps, rest times, and equipment adjustments. The app also delivers progress summaries and real-time coaching that explains why changes are suggested. Journaling app Stoic now provides prompts tailored to a user’s mood or sleep patterns, and generates compassionate messages or reflections directly from entries without sending data off-device.
Other apps are adopting similar approaches. SwingVision analyzes tennis and pickleball games to provide targeted improvements, while 7 Minute Workout creates dynamic exercise sessions based on user needs, such as injury avoidance. Gratitude generates affirmations from journal entries, and Wakeout! builds personalized micro-breaks with contextual explanations.
In education, CellWalk uses the framework to give conversational explanations of biological terms inside its 3D cell explorer. Explanations adapt to each learner’s profile, making complex science more approachable. These examples highlight Apple’s push to empower developers with native AI tools, contrasting with competitor approaches that rely more heavily on cloud-based models.
By making its foundation models available as part of Apple Intelligence, Apple is ensuring that apps can deliver intelligent features without compromising privacy or requiring subscriptions to external AI services. This positions iOS 26 and its ecosystem as a leader in user-friendly, integrated AI, while also laying groundwork for developers to experiment with new intelligent app designs.
Check out other iOS 26 apps that are using local AI models on device.
