PyTorch is a Python package that provides two high-level features:

• Tensor computation (like NumPy) with strong GPU acceleration
• Deep neural networks built on a tape-based autograd system

You can reuse your favorite Python packages such as NumPy, SciPy, and Cython to extend PyTorch when needed.

Our trunk health (Continuous Integration signals) can be found at hud.pytorch.org.

• More About PyTorch
  • A GPU-Ready Tensor Library
  • Dynamic Neural Networks: Tape-Based Autograd
  • Python First
  • Imperative Experiences
  • Fast and Lean
  • Extensions Without Pain
• Installation
  • Binaries
    • NVIDIA Jetson Platforms
  • From Source
    • Prerequisites
      • NVIDIA CUDA Support
      • AMD ROCm Support
      • Intel GPU Support
    • Get the PyTorch Source
    • Install Dependencies
    • Install PyTorch
      • Adjust Build Options (Optional)
  • Docker Image
    • Using pre-built images
    • Building the image yourself
  • Building the Documentation
    • Building a PDF
  • Previous Versions
• Getting Started
• Resources
• Communication
• Releases and Contributing
• The Team
• License

Learn the basics of PyTorch

At a granular level, PyTorch is a library that consists of the following components:

Usually, PyTorch is used either as:

• A replacement for NumPy to use the power of GPUs.
• A deep learning research platform that provides maximum flexibility and speed.

Elaborating Further:

If you use NumPy, then you have used Tensors (a.k.a. ndarray).

PyTorch provides Tensors that can live either on the CPU or the GPU and accelerates the computation by a huge amount.

We provide a wide variety of tensor routines to accelerate and fit your scientific computation needs such as slicing, indexing, mathematical operations, linear algebra, reductions. And they are fast!

PyTorch has a unique way of building neural networks: using and replaying a tape recorder.

Most frameworks such as TensorFlow, Theano, Caffe, and CNTK have a static view of the world. One has to build a neural network and reuse the same structure again and again. Changing the way the network behaves means that one has to start from scratch.

With PyTorch, we use a technique called reverse-mode auto-differentiation, which allows you to change the way your network behaves arbitrarily with zero lag or overhead. Our inspiration comes from several research papers on this topic, as well as current and past work such as torch-autograd, autograd, Chainer, etc.

While this technique is not unique to PyTorch, it's one of the fastest implementations of it to date. You get the best of speed and flexibility for your crazy research.

PyTorch is not a Python binding into a monolithic C++ framework. It is built to be deeply integrated into Python. You can use it naturally like you would use NumPy / SciPy / scikit-learn etc. You can write your new neural network layers in Python itself, using your favorite libraries and use packages such as Cython and Numba. Our goal is to not reinvent the wheel where appropriate.

PyTorch is designed to be intuitive, linear in thought, and easy to use. When you execute a line of code, it gets executed. There isn't an asynchronous view of the world. When you drop into a debugger or receive error messages and stack traces, understanding them is straightforward. The stack trace points to exactly where your code was defined. We hope you never spend hours debugging your code because of bad stack traces or asynchronous and opaque execution engines.

PyTorch has minimal framework overhead. We integrate acceleration libraries such as Intel MKL and NVIDIA (cuDNN, NCCL) to maximize speed. At the core, its CPU and GPU Tensor and neural network backends are mature and have been tested for years.

Hence, PyTorch is quite fast — whether you run small or large neural networks.

The memory usage in PyTorch is extremely efficient compared to Torch or some of the alternatives. We've written custom memory allocators for the GPU to make sure that your deep learning models are maximally memory efficient. This enables you to train bigger deep learning models than before.

Writing new neural network modules, or interfacing with PyTorch's Tensor API was designed to be straightforward and with minimal abstractions.

You can write new neural network layers in Python using the torch API or your favorite NumPy-based libraries such as SciPy.

If you want to write your layers in C/C++, we provide a convenient extension API that is efficient and with minimal boilerplate. No wrapper code needs to be written. You can see a tutorial here and an example here.

Commands to install binaries via Conda or pip wheels are on our website: https://pytorch.org/get-started/locally/

Python wheels for NVIDIA's Jetson Nano, Jetson TX1/TX2, Jetson Xavier NX/AGX, and Jetson AGX Orin are provided here and the L4T container is published here

They require JetPack 4.2 and above, and @dusty-nv and @ptrblck are maintaining them.

If you are installing from source, you will need:

• Python 3.9 or later
• A compiler that fully supports C++17, such as clang or gcc (gcc 9.4.0 or newer is required, on Linux)
• Visual Studio or Visual Studio Build Tool (Windows only)

* PyTorch CI uses Visual C++ BuildTools, which come with Visual Studio Enterprise, Professional, or Community Editions. You can also install the build tools from https://visualstudio.microsoft.com/visual-cpp-build-tools/. The build tools do not come with Visual Studio Code by default.

An example of environment setup is shown below:

• Linux:

$ source <CONDA_INSTALL_DIR>/bin/activate
$ conda create -y -n <CONDA_NAME>
$ conda activate <CONDA_NAME>

• Windows:

$ source <CONDA_INSTALL_DIR>\Scripts\activate.bat
$ conda create -y -n <CONDA_NAME>
$ conda activate <CONDA_NAME>
$ call "C:\Program Files\Microsoft Visual Studio\<VERSION>\Community\VC\Auxiliary\Build\vcvarsall.bat" x64

A conda environment is not required. You can also do a PyTorch build in a standard virtual environment, e.g., created with tools like uv, provided your system has installed all the necessary dependencies unavailable as pip packages (e.g., CUDA, MKL.)

If you want to compile with CUDA support, select a supported version of CUDA from our support matrix, then install the following:

• NVIDIA CUDA
• NVIDIA cuDNN v8.5 or above
• Compiler compatible with CUDA

Note: You could refer to the cuDNN Support Matrix for cuDNN versions with the various supported CUDA, CUDA driver, and NVIDIA hardware.

If you want to disable CUDA support, export the environment variable USE_CUDA=0. Other potentially useful environment variables may be found in setup.py. If CUDA is installed in a non-standard location, set PATH so that the nvcc you want to use can be found (e.g., export PATH=/usr/local/cuda-12.8/bin:$PATH).

If you are building for NVIDIA's Jetson platforms (Jetson Nano, TX1, TX2, AGX Xavier), Instructions to install PyTorch for Jetson Nano are available here

If you want to compile with ROCm support, install

• AMD ROCm 4.0 and above installation
• ROCm is currently supported only for Linux systems.

By default the build system expects ROCm to be installed in /opt/rocm. If ROCm is installed in a different directory, the ROCM_PATH environment variable must be set to the ROCm installation directory. The build system automatically detects the AMD GPU architecture. Optionally, the AMD GPU architecture can be explicitly set with the PYTORCH_ROCM_ARCH environment variable AMD GPU architecture

If you want to disable ROCm support, export the environment variable USE_ROCM=0. Other potentially useful environment variables may be found in setup.py.