Install TensorFlow 2

TensorFlow is tested and supported on the following 64-bit systems:

Python 3.5–3.7
Ubuntu 16.04 or later
Windows 7 or later

macOS 10.12.6 (Sierra) or later (no GPU support)
Raspbian 9.0 or later

# Requires the latest pip
pip install --upgrade pip

# Current stable release for CPU and GPU
pip install tensorflow

# Or try the preview build (unstable)
pip install tf-nightly

Download a package

Install TensorFlow with Python's pip package manager.

TensorFlow 2 packages require a pip version >19.0.

Official packages available for Ubuntu, Windows, macOS, and the Raspberry Pi.

See the GPU guide for CUDA®-enabled cards.

Read the pip install guide

Run a TensorFlow container

The TensorFlow Docker images are already configured to run TensorFlow. A Docker container runs in a virtual environment and is the easiest way to set up GPU support.

 docker pull tensorflow/tensorflow:latest-py3  # Download latest stable image
 docker run -it -p 8888:8888 tensorflow/tensorflow:latest-py3-jupyter  # Start Jupyter server

Read the Docker install guide

Google Colab: An easy way to learn and use TensorFlow

No install necessary—run the TensorFlow tutorials directly in the browser with Colaboratory, a Google research project created to help disseminate machine learning education and research. It's a Jupyter notebook environment that requires no setup to use and runs entirely in the cloud. Read the blog post.

Build your first ML app

Create and deploy TensorFlow models on web and mobile.

Web developers

TensorFlow.js is a WebGL accelerated, JavaScript library to train and deploy ML models in the browser and for Node.js.

Mobile developers

TensorFlow Lite is a lightweight solution for mobile and embedded devices.