TensorFlow is back at Google I/O on May 14! Register now

Deploy machine learning models on mobile devices and edge devices

TensorFlow Lite is a mobile terminal library, which can be used to deploy models on mobile devices, microcontrollers and other edge devices.

View Guide

The guide introduces the concepts and components of TensorFlow Lite.

View example

Explore Android and iOS applications using TensorFlow Lite.

View Tutorials

Learn how to use TensorFlow Lite for common use cases.

Select Model

Select a new model or retrain an existing model.

Read the developer guide

transformation

Use the TensorFlow Lite Converter to convert the TensorFlow model to a compressed planar buffer.

Read the developer guide

deploy

Get the compressed. tflite file and load it into the mobile device or embedded device.

Read the developer guide

optimization

Quantize by converting 32-bit floating point numbers to more efficient 8-bit integers, or run on the GPU.

Read the developer guide

image classification

Identify hundreds of objects, including people, activities, animals, plants and places.

Object detection

Use bounding boxes to detect multiple objects. Yes, it can also detect dogs and cats.

Question answer

Use the advanced natural language model to answer questions according to the content of a given text paragraph through BERT.

TensorFlow Lite on GitHub

Community Forum

Ask questions on Stack Overflow

Explore Dev Library Community Project

Google Photos

Gboard

Gmail

Google Assistant

VSCO

Shazam

Nest

Lose it

Deploy machine learning models on mobile devices and edge devices

working principle

Select Model

transformation

deploy

optimization

Solutions to common problems

Community participation

Some of our TensorFlow Lite users

News and announcements

Start using TensorFlow Lite