lucidrains/vit-pytorch Alternatives

Official JAX implementation; choose it when training on TPU with JAX instead of PyTorch.

Full CV toolkit; choose it when you need a broad model zoo and training pipeline, not just ViT.

An all-in-one toolkit for computer vision

JAX CV research library; choose it when your stack is JAX and you need production-grade ViT/other models.

Scenic: A Jax Library for Computer Vision Research and Beyond

Specific Swin Transformer implementation; choose it when you want Swin's shifted-window attention and official pretrained weights.

This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".

Hybrid Mamba-Transformer backbone; choose it when you need a modern, efficient alternative to ViT with linear complexity.

[CVPR 2025] Official PyTorch Implementation of MambaVision: A Hybrid Mamba-Transformer Vision Backbone

Large model collection; choose it when you want pretrained weights and standardized training/inference across many backbones.

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

Official JAX codebase for ViT; choose it for large-scale distributed training and reproducing research results.

Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

Config-driven OpenMMLab toolbox; choose it when you want modular training/config system and many pretrained models.

OpenMMLab Pre-training Toolbox and Benchmark

Apple's CV training library; choose it when you prefer a self-contained library with efficient model implementations and quantization.

CVNets: A library for training computer vision networks

PoolFormer implementation; choose it when you need a simple non-attention metaformer baseline.

PoolFormer: MetaFormer Is Actually What You Need for Vision (CVPR 2022 Oral)