Check its license. It has its own license: read it before using it for work.

Is pytorch still maintained?

Yes, active. Last change today. Latest release v2.15.0, 14 days ago.

Widely used. 105k stars, 10 years old, by an organization. As with any download, get it from the official repo.

How to install on Linux

pytorch/pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

PythonC/C++ (CMake)#autograd#deep-learning#gpu#machine-learning

Before you start

Is it free?: Check its licenseIt has its own license: read it before using it for work.
Still maintained?: Yes, activeLast change today. Latest release v2.15.0, 14 days ago.
Can I trust it?: Widely used105k stars, 10 years old, by an organization. As with any download, get it from the official repo.

Difficulty: Moderate
Time: about 15 min
Stars: ★ 105,389
Activity: Updated this month

Saved setup from 5 h ago

Will it work on Linux?

Yes, with Docker on Linux

It's a self-hosted app: the README runs it with Docker, and you open it in your browser.

Your system

macOS Windows Linux

Start here

Website ↗pytorch.orgThe project's own site.

Run it with Docker

A self-hosted app: Docker runs it on your computer and you open it in the browser. Needs Docker Desktop (or Docker Engine on Linux). Run docker compose commands inside the cloned repo folder.

Docker

docker run --gpus all --rm -ti --ipc=host pytorch/pytorch:latest

From the README, exactly as the maintainers wrote it.

Did you get it running?

No reports for Linux yet.

Still stuck? Get a written guide

GitManual AI Agent reads the 15 key files and writes a step-by-step guide for Linux. Once written, it's saved for everyone.

Parts list

1×Python >=3.11runtime
1×C/C++ compiler runtime
1×CMake 3.27runtime
1×pip package manager

Share this manual

Is this your repo?

Add a badge to your README so newcomers find these steps.

[![Setup manual](https://gitmanual.com/pytorch/pytorch/badge.svg)](https://gitmanual.com/pytorch/pytorch)

Good to know

Native projects often need extra system libraries. If the build stops with a missing package, check the maintainers' build guide (see Start here).
The README mentions NVIDIA CUDA. Check its hardware section: some features (or all of them) may need an NVIDIA GPU.
Dockerfile available (base image: ${BASE_IMAGE}).
Has AGENTS.md — instructions for coding agents. You can open Claude Code (or another agent) in the cloned folder and ask it to do the setup for you.
Nested projects: .ci/pytorch/test_example_code/CMakeLists.txt, .devcontainer/cuda/requirements.txt, .github/actions/pi-agent/package.json, android/pytorch_android/CMakeLists.txt and 11 more

Details

License: Not detected
Open issues: 18,178
Branch: main

Files we read (15)

.github/workflows/_binary-build-flash-attention-wheel-linux.yml

.github/workflows/_binary-build-flash-attention-wheel-windows.yml

.github/workflows/_binary-build-linux.yml

.github/workflows/_binary-test-linux.yml

.github/workflows/_binary-test-macos.yml

.github/workflows/_binary-test-rocm-linux.yml

.gitmodules

CMakeLists.txt

CONTRIBUTING.md

Dockerfile

Makefile

README.md

pyproject.toml

requirements.txt

setup.py

22,436 files in the repo

Yes, with Docker on Linux

Start here

Run it with Docker

Did you get it running?

Still stuck? Get a written guide

More like pytorch that run on Linux