Explore and extend models from the latest cutting edge research.
Discover and publish models to a pre-trained model repository designed for research exploration. Check out the models for Researchers, or learn How It Works. Contribute Models.
*This is a beta release – we will be collecting feedback and improving the PyTorch Hub over the coming months.
AlexNet
The 2012 ImageNet winner achieved a top-5 error of 15.3%, more than 10.8 percentage points lower than that of the runner up.
Densenet
Dense Convolutional Network (DenseNet), connects each layer to every other layer in a feed-forward fashion.
GoogLeNet
GoogLeNet was based on a deep convolutional neural network architecture codenamed “Inception” which won ImageNet 2014.
GPUNet
GPUNet is a new family of Convolutional Neural Networks designed to max out the performance of NVIDIA GPU and TensorRT.
EfficientNet
EfficientNets are a family of image classification models, which achieve state-of-the-art accuracy, being an order-of-magnitude smaller and faster. Trained with mixed precision using Tensor Cores.
ResNeXt101
ResNet with bottleneck 3×3 Convolutions substituted by 3×3 Grouped Convolutions, trained with mixed precision using Tensor Cores.
SE-ResNeXt101
ResNeXt with Squeeze-and-Excitation module added, trained with mixed precision using Tensor Cores.
Silero Speech-To-Text Models
A set of compact enterprise-grade pre-trained STT Models for multiple languages.
Silero Text-To-Speech Models
A set of compact enterprise-grade pre-trained TTS Models for multiple languages
Once-for-All
Once-for-all (OFA) decouples training and search, and achieves efficient inference across various edge devices and resource constraints.
U-Net for brain MRI
U-Net with batch normalization for biomedical image segmentation with pretrained weights for abnormality segmentation in brain MRI
Semi-supervised and semi-weakly supervised ImageNet Models
ResNet and ResNext models introduced in the “Billion scale semi-supervised learning for image classification” paper
SimpleNet
Lets Keep it simple, Using simple architectures to outperform deeper and more complex architectures
