Apple and Google Team Up on Foundation Models, Free for Small Developers
Apple's next-gen Foundation Models were built in collaboration with Google Gemini. Small Business Program developers get free cloud API access. Core AI framework lets you run full LLMs locally on Apple Silicon.
In a notable collaboration, Apple revealed at WWDC 2026 that its next-generation Foundation Models were built in partnership with Google and Gemini models. The models power on-device and server-side AI features across Apple platforms, and small developers get free cloud API access.
The Apple-Google AI stack
Apple's Foundation Models framework now serves as a single native Swift API that supports:
- On-device models with image input, running locally on Apple Silicon
- Server models via Private Cloud Compute, Apple's privacy-preserving cloud
- Custom skills, developers define model capabilities declaratively
- Dynamic Profiles, update how models interact with apps at runtime
- Third-party models, Claude, Gemini, or any provider implementing Apple's new language model protocol
Core AI: run LLMs locally
For developers wanting to ship their own custom models, Core AI is a new framework optimized for Apple Silicon's unified memory and Neural Engine. It's designed to be the best way to run full-scale LLMs on-device, no cloud dependency, no API costs, complete privacy.
Free for small developers
Developers enrolled in the App Store Small Business Program (under 2 million first-time downloads) get access to Apple Foundation Models running on Private Cloud Compute at no cloud API cost. This is a strategic play to keep indie developers building for Apple platforms rather than defecting to web or cross-platform alternatives where AI APIs are pay-per-token.
The collaboration angle
Apple building models with Google rather than competing alone signals a pragmatic shift. Apple brings on-device optimization and privacy infrastructure; Google brings model training expertise at scale. The result: Apple developers get state-of-the-art models without Apple having to match OpenAI or Anthropic's training investments alone.
The bottom line
Apple's AI strategy is now clear: partner for model training, own the on-device runtime, and give small developers free access to keep the ecosystem growing. If you're an indie Apple developer, you just got access to production LLMs at zero incremental cost.