README.md

metadata

license: other
library_name: executorch
tags:
  - android
  - ios
  - on-device
  - pytorch
  - react-native

Llama-3.2-1B-Executorch-SpinQuant

This repository contains the llama3_2_1b_spinquant.pte model, exported for use with ExecuTorch.

Details

• Model: Llama 3.2 Instruct
• Format: .pte (ExecuTorch)
• Quantization: Llama 3.2 1B Instruct model exported for ExecuTorch with SpinQuant (4-bit). Compatible with React Native.

Usage

This model is ready to be used in mobile applications (iOS/Android) via the ExecuTorch runtime or react-native-executorch.

1. Download tokenizer.model and llama3_2_1b_spinquant.pte.
2. Place them in your app's asset folder.