On-device AI for React Native.

Generate text, transcribe speech, work with images, and search by meaning. Runs on iOS and Android. No API key.

Use it in an Expo app or a React Native app with Expo modules installed. You need a native build; Expo Go is not supported.

npx expo install expo-ai-kit

All features are off by default. Turn on the ones you use in the config plugin, then make a native build.

Installation and first run →

What do you want to build?

Capability	Use it for
LLM	Chat, text generation, structured JSON, and tool calling
Speech	Live dictation and audio transcription
Vision	Background removal, image labels, OCR, and face detection
Embeddings	Semantic search over your own data

A text request

import { isAvailable, prepareBuiltInModel, sendMessage } from 'expo-ai-kit';

export async function ask(question: string) {
  if (!(await isAvailable())) return 'Built-in text model unavailable.';
  await prepareBuiltInModel();
  const { text } = await sendMessage([{ role: 'user', content: question }]);
  return text;
}

Support depends on the device and feature. The built-in text model needs Apple Intelligence on iOS 26+ or a supported Android device. Other features have their own requirements. See platform support or use a downloadable model.

Already using the Vercel AI SDK? Import the provider from expo-ai-kit/ai.