# software-mansion/react-native-executorch

Declarative way to run AI models in React Native on device, powered by ExecuTorch.

Repository: https://github.com/software-mansion/react-native-executorch
Canonical: https://ross.abutalabs.com/products/react-native-executorch
Homepage: http://docs.swmansion.com/react-native-executorch/
Language: C++
License: NOASSERTION
License Family: other
Topics: computer-vision, executorch, llm-inference, object-detection, ocr, on-device-ai, react-native-ai, speech-to-text, text-to-speech, image-embeddings, segmentation, text-embeddings, vlm
Last push: 2026-08-26T19:36:17+00:00

## Health v2 (maintenance only)
Score: 88/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 96, longevity 49
- inputs: {"age_days": 686, "days_push": 7, "days_rel": 30, "gap_med": 7.0, "n_releases_24m": 49}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1702, forks 95 (observed 2026-08-28T04:05:24.382185+00:00)

## What it is
React Native ExecuTorch is a declarative React Native library for running AI models on-device, powered by Meta's ExecuTorch runtime. It ships pre-exported models for LLMs, vision-language models, computer vision, speech-to-text, text-to-speech, OCR, and embeddings, all running locally without cloud API calls.

## Use cases
- run LLMs locally in a React Native app
- on-device object detection and image segmentation
- offline speech-to-text and text-to-speech in mobile apps
- generate image and text embeddings for local semantic search or RAG
- OCR a photo of a document and feed the text to a local LLM
- build a private, offline AI chat assistant in Expo
- image classification without a cloud API

## When to choose
- you build React Native or Expo apps and want AI inference fully on-device for privacy or offline use
- you want pre-packaged models (Llama, Phi, Whisper, Kokoro, YOLO, SAM) with a simple declarative JS API
- you want to avoid cloud inference costs or send user data to servers

## When to avoid
- your app uses the old React Native architecture (New Architecture is required)
- you need server-side or GPU-scale inference rather than mobile on-device models
- you don't use React Native or JavaScript as your app layer

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-inference, computer-vision, speech-recognition, tts, ocr, rag
- domain: machine-learning, artificial-intelligence, mobile-development, computer-vision, speech-processing, large-language-models
- platform: cross-platform
- tags: react-native, executorch, on-device-ai, expo, embeddings, vlm, segmentation, object-detection, offline-ai, android, ios, mobile

## Member repositories
- software-mansion/react-native-executorch (main) score 88

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:24.382185+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:37:41.192343+00:00, confidence not recorded.
  - readme: https://github.com/software-mansion/react-native-executorch (fetched 2026-08-28T04:05:24.382185+00:00, sha 00ef0dc3af64)
  - homepage: http://docs.swmansion.com/react-native-executorch/ (fetched 2026-08-29T11:12:07.085170+00:00, sha 5d1ce2017eed)
  - site_page: https://docs.swmansion.com/react-native-executorch/docs/fundamentals/getting-started (fetched 2026-08-29T11:12:07.090800+00:00, sha 26b3ccff02da)
- Data as of 2026-08-30T08:39:29.467469+00:00.