OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

React Native Executorch

React Native ExecuTorch is a TypeScript library that lets React Native applications run machine-learning inference on-device through the ExecuTorch runtime, using pre-exported or custom .pte models.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

React Native Executorch — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when a React Native app needs model inference on the device, including workflows designed to continue offline after model download.

Who it is for

  • React Native developers adding local model inference to iOS or Android applications.
  • Teams building mobile features for vision, language, speech synthesis or transcription.
  • Developers who need task hooks for common models or want to compose custom TypeScript inference pipelines.

Documented uses

  • Add image classification or object detection to a mobile app.
  • Build an on-device language-model assistant.
  • Add local text-to-speech or speech transcription.
  • Create custom inference pipelines using TypeScript and lower-level runtime operations.
Before you choose

Requirements & limits

The repository is public and identifies the project as open source, but the supplied repository data does not include license details. The README lists no service account requirement; its quickstart describes installation as a project dependency.

  • Requires React Native's New Architecture and React Native 0.83+ or Expo SDK 55+ Development Builds; Expo Go is unsupported.
  • Requires iOS 17+ or Android 13+ (minSdkVersion 26), plus a supported react-native-worklets version and Babel plugin configuration.
  • Pre-exported model downloads send an anonymous download event, with an opt-out method named in the README.
  • Documented acceleration backends include XNNPACK, Core ML, MLX and Vulkan; available execution paths depend on the platform.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the library and add a language-model chat hook

Set up React Native ExecuTorch and follow the documented quickstart pattern for a language-model session.

Before you start

  • React Native 0.83+ or Expo SDK 55+ with a Development Build; Expo Go is unsupported.
  • New Architecture enabled, iOS 17+ or Android 13+ (minSdkVersion 26).
  • react-native-worklets >=0.10.0 and <0.13.0; configure react-native-worklets/plugin as the last Babel plugin.
  • A package manager and the listed peer dependencies: react-native-worklets and react-native-blob-util.

Steps

  1. Install react-native-executorch, react-native-worklets and react-native-blob-util using npm, Yarn or pnpm.

  2. Configure the required React Native architecture, platform minimums and Babel plugin order.

  3. Import models and useLLMChatSession from react-native-executorch, then create a session with models.llm.LFM2_5_1_2B.DEFAULT.

  4. Follow the quickstart pattern: wait until the session is ready, call sendMessage with a prompt, and read the returned messages.

Expected outcome

A React Native component wired to the documented language-model chat session pattern.

Guide limitations

  • The quickstart shows the integration pattern but does not establish performance or compatibility for a particular device or model.
  • Pre-exported model downloads send an anonymous event; the README identifies setTelemetryEnabled(false) as the opt-out.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
software-mansion/react-native-executorch · Official homepage title · retained in its source language
How we review sourcesSuggest a correction