ConvNeXt

Category: CNN · Code: timm/models/convnext.py · Architectures: 31 · Pretrained weights: 101 (61 with ImageNet-1k results)

ConvNeXt is a pure convolutional network that adopts design choices from vision transformers, such as large depthwise kernels, inverted bottlenecks and LayerNorm. ConvNeXt-V2 adds Global Response Normalization (GRN) and is pretrained as a fully convolutional masked autoencoder (FCMAE).

Papers

• A ConvNet for the 2020s
• ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders
• Reproducible scaling laws for contrastive language-image learning
• DINOv3
• Efficient Universal Perception Encoder

Usage

See the Quickstart for loading pretrained weights, image preprocessing and fine-tuning, and Feature Extraction for using the models as backbones.

>>> import timm
>>> model = timm.create_model('convnext_tiny.in12k_ft_in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)

Model API support

✓ supported, ✓ (square) works for square sizes only, ✗ not supported, – not needed (the model already accepts any input size). Checked by running each feature on a randomly initialized model (ConvNeXt with convnext_zepto_rms at 224), other architectures of the same class are expected to behave the same. See the Model API page for details.

Architectures

GMACs and parameter counts are for the default input size and classifier of each architecture, from the benchmark results.

Pretrained weights

Top-1 is ImageNet-1k validation accuracy from results-imagenet.csv at the listed eval size, - if the weight has no ImageNet-1k result (e.g. pretraining or image-text weights). Classes is the size of the pretrained classifier, none for weights without one.

Weights are grouped by source: the paper or origin given in the pretrained config, else a Hub organization other than timm, else the first part of the pretrained tag (architecture.tag). Sources with a single weight are listed under other.

ConvNeXt V2

Pretrained configs reference ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders.

clip tags

fb tags

d1 tags

dinov3 tags

eupe tags

d1h tags

ra4 tags

Other

Citation

@Article{liu2022convnet,
  author  = {Zhuang Liu and Hanzi Mao and Chao-Yuan Wu and Christoph Feichtenhofer and Trevor Darrell and Saining Xie},
  title   = {A ConvNet for the 2020s},
  journal = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year    = {2022},
}

@article{Woo2023ConvNeXtV2,
  title={ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders},
  author={Sanghyun Woo, Shoubhik Debnath, Ronghang Hu, Xinlei Chen, Zhuang Liu, In So Kweon and Saining Xie},
  year={2023},
  journal={arXiv preprint arXiv:2301.00808},
}