#2424th of 26 in ML frameworks and libraries
vit-pytorch
PyTorch implementations of Vision Transformer and many of its variants
- Stars
- 25.5k
- License
- MIT
- Last commit
- Sep 2026
- Last release
- Feb 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
A pip-installable PyTorch library with compact implementations of ViT and dozens of variants, including SimpleViT, NaViT, CaiT, CrossViT, MaxViT and MobileViT, plus masked autoencoder and DINO wrappers. Each model is a plain nn.Module built from hyperparameters, with a short usage snippet in the README. It ships no pretrained weights; the README points to timm for those.
Who it is for: Researchers and engineers who want readable ViT code to train or modify
Strengths
- Covers many ViT variants (NaViT, CaiT, CrossViT, LeViT, CvT, MaxViT) with similar constructors
- Includes distillation, masked autoencoder and DINO wrappers for training
- Installs with pip install vit-pytorch; MIT licensed
- Each model has a runnable snippet showing input and output shapes
Weaknesses
- No pretrained weights; README refers to timm for pretrained models
- No training scripts, datasets or benchmark results in the README
- README is a list of snippets; no API reference or docs site
- NaViT nested-tensor variant requires PyTorch 2.5
What it needs
- GPU optional
- Needs PyTorch, torchvision
- Models: ViT, SimpleViT, NaViT, CaiT, CrossViT
Also in ML frameworks and libraries
See all 26| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 89 out of 100 |
| #4 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 85 out of 100 |
| #5 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 83 out of 100 |