pytorch-image-models
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
About pytorch-image-models
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
What you should know about pytorch-image-models
pytorch-image-models — 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. It is categorized under AI Tools and primarily built with Python. The project has gathered 36,651 stars and 5,144 forks on GitHub, indicating strong adoption among developers.
Pricing & licensing: This tool is offered free of charge , released under the Apache-2.0 license. The source code is openly available on GitHub, allowing engineers to audit, contribute, or fork as needed.
Use cases & topics: pytorch-image-models is associated with the following topics: augmix, convnext, distributed-training, efficientnet, image-classification, imagenet, maxvit, mixnet. Teams working in augmix / convnext / distributed-training spaces typically evaluate this kind of tool when scoping new architecture decisions or replacing legacy components.
Getting started: Check out the official GitHub repository for installation steps, configuration examples, and the latest release notes. Most teams hit value within the first week if the tool aligns with their existing AI Tools stack.
Editor's note from Fanny Engriana (Founder, Wardigi Digital Agency): when evaluating tools in the AI Tools category for our agency clients, we look at three things first — license clarity, community size, and active maintenance. Tools with explicit license terms and ongoing commits tend to remain viable across multi-year projects.