#212212th of 242 in ML frameworks and libraries
pytorchvideo
PyTorch library of video models, datasets and transforms for research
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 3.6k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Sep 2021
- Language
- Python
3,534 on 25 Apr 20263,569 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
PyTorchVideo is a PyTorch library for video understanding research. It provides a model zoo of pretrained video models (including X3D, SlowFast and Multiscale Vision Transformers), data loaders for video datasets, and video-specific transforms. It installs with pip and includes tutorials and a Hugging Face Spaces demo of X3D.
Who it is for: Researchers and engineers training or running video models in PyTorch
Strengths
- Model zoo with pretrained video models and published benchmarks
- Modular components built on PyTorch, so they fit existing PyTorch code
- Data loaders for several video datasets plus video-specific transforms
- Installs with pip install pytorchvideo; tutorials included
Weaknesses
- Last PyPI release was 2021-09-10, despite later commits
- Research-oriented; no serving layer or application UI
- README lists only a 2021 update, so newer model coverage is unclear
- No Docker image provided
What it needs
- Needs PyTorch, Python >=3.7
- Models: X3D, SlowFast, Multiscale Vision Transformers
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in ML frameworks and libraries
See all 242| Rank | Project | Score |
|---|---|---|
| 1 | transformersModel-definition library for pretrained text, vision, audio and multimodal models | 94 out of 100 |
| 2 | ultralyticsPython library and CLI for training and running YOLO vision models | 94 out of 100 |
| 3 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 94 out of 100 |
| #4 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |