Getting Started

Use the AI that ships with iOS: Apple Intelligence for text, plus speech recognition, speech synthesis and embeddings. The models are part of the system and run on the device, so there’s nothing to bundle. iOS downloads some assets on first use.

Feature	Model	Apple API
Text generation	apple()	Foundation Models
Embeddings	apple.embeddingModel()	NLContextualEmbedding
Transcription	apple.transcriptionModel()	SpeechAnalyzer
Speech	apple.speechModel()	AVSpeechSynthesizer

Installation

npm

npm install @react-native-ai/apple ai

yarn

yarn add @react-native-ai/apple ai

pnpm

pnpm add @react-native-ai/apple ai

bun

bun add @react-native-ai/apple ai

The provider works with AI SDK v7 and needs its polyfills. Examples with schemas also use zod.

Requirements

• React Native New Architecture
• iOS 26 for text generation and transcription, on a device with Apple Intelligence turned on. Embeddings and speech work on older versions.
• Xcode 27 to build with the iOS 27 features. Older SDKs leave them out, see iOS Versions.

Usage

import { apple } from '@react-native-ai/apple'
import { generateText } from 'ai'

if (apple.isAvailable()) {
  const { text } = await generateText({
    model: apple(),
    prompt: 'Explain quantum computing in simple terms.',
  })
}

isAvailable() is false when the device is not eligible, Apple Intelligence is off, or the model is still downloading. getModelInfo() tells you which. Before iOS 26, isAvailable() is false and getModelInfo() rejects with UNSUPPORTED_OS.

To test in the iOS Simulator, turn on Apple Intelligence on your Mac first, see Running on Simulator.