Ship Onde from React Native.

Keep your app in JavaScript or TypeScript. Put inference in the native layer where it belongs. That is the promise of @ondeinference/react-native.

View source All SDKs

npm package

Published as `@ondeinference/react-native`. Install it with the package manager your app already uses and keep moving.

Native engine

The JavaScript layer stays thin on purpose. The heavy lifting happens in the native Onde engine, where on-device inference actually belongs.

Expo-friendly

The package is built as an Expo module, which makes it a better fit for modern React Native apps than a random pile of bridge code.

Small surface, real app

Load a model, send a message, stream results, manage history. The API stays compact because the point is to ship, not to admire the wrapper.

Add the package and go.

Install from npm, pnpm, or yarn. If you want to test model downloads or GGUF export before you ship the app build, use Onde CLI. If you want the broader case for local inference, read The Latency War.

# npm npm install @ondeinference/react-native # pnpm pnpm add @ondeinference/react-native # yarn yarn add @ondeinference/react-native

Load. Prompt. Render.

The JavaScript API stays small because it should. Load the default model, send a message, and render the result in your app.

import { OndeChatEngine } from '@ondeinference/react-native'; const engine = new OndeChatEngine(); await engine.loadDefaultModel({ systemPrompt: 'You are a helpful assistant.', }); const result = await engine.sendMessage({ message: 'Hello!', }); console.log(result.text);

Ship Onde from React Native.

Add the package and go.

Load. Prompt. Render.

Mobile first, for real.