Bringing the latest Gemini models to Apple developers

TL;DR

• 1

  Apple developers can now call Gemini through Apple’s native Foundation Models framework — Starting with iOS 27, macOS 27, iPadOS 27, visionOS 27, and watchOS 27, Gemini is exposed via the new public LanguageModel protocol through the Firebase Apple SDK.

• 2

  Gemini and Apple’s on-device model now sit behind the same API surface — If you already use Foundation Models, switching to Gemini is basically a small code change—swap the model instance—so you can move between local and cloud inference based on cost, latency, and agentic app needs.

• 3

  Firebase AI Logic removes the usual backend work — Google says developers can add the latest Gemini models directly into iOS, macOS, iPadOS, and visionOS apps without building and maintaining a separate server, while Firebase App Check helps protect those model-access APIs from abuse.

• 4

  Gemini is now built into Xcode for multi-step coding work — After onboarding through Xcode’s Intelligence settings panel, developers can use Gemini natively to review code, fix bugs, and build features without bouncing between tools or windows.

• 5

  Google is offering both solo-dev and enterprise auth paths — Individual developers can use a self-serve Gemini API key from Google AI Studio with free and paid tiers, while enterprises can use Gemini Enterprise Agent Platform for dedicated corporate quotas and data privacy controls.