---
license: apache-2.0
---

# ConvNext

ConvNext model pre-trained on ImageNet-21k (14 million images, 21,843 classes) at resolution 224x224. It was introduced in the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545).
The weights were converted from the `convnext_large_22k_224.pth` file presented in the [official repository](https://github.com/facebookresearch/ConvNeXt).