Project AURIONAI Use-Cases Research & Innovation
Meta
visionfamily

ConvNeXt

Meta family of modernised pure convolutional networks that match vision transformers on ImageNet, COCO and ADE20K.

MetaMetaUSA
Diagram comparing the Swin Transformer, ResNet and ConvNeXt blocks, from the ConvNeXt model card.
Levelfamily
Classvision
Parametersnot in the notes
Contextnot in the notes
Released10 Jan 2022
Opennessunknown
Runs onmobile or NPU, CPU server
Statusactive

Lineage

MetaConvNeXt V2 belongs to this familyvision2 Jan 2023197.96M
MetaConvNeXt belongs to this familyvision10 Jan 2022
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. Confidence and evidence are the notes' own.

R&D Topics

Image classification backbone architectures (CNN and vision transformer)produced by

Identity

Developer
Meta
Family
not in the notes
Released
10 Jan 2022 sourcesannouncement page
Weights published
–
Record checked
12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
CNN
Total parameters
– sources
Active parameters
–
Checkpoint parameters
–
Note
ConvNeXt T, S, B, L, XL; ConvNeXt V2 from Atto to Huge.
Context length
not in the notes
Input
image
Output
labels and boxes
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
unknown
Licence
not in the notes sources
Access routes
open weights download, cloud marketplace
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
mobile or NPU, CPU server
Footprint
Edge or toolkit builds listed in: Qualcomm AI Hub, OpenVINO.
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. arxiv.org/abs/2201.03545 https://arxiv.org/abs/2201.03545
  2. ai.azure.com/catalog/models?task=image-classification https://ai.azure.com/catalog/models?task=image-classification
  3. huggingface.co/facebook/convnext-large-224 https://huggingface.co/facebook/convnext-large-224
  4. aihub.qualcomm.com/models/convnext_base https://aihub.qualcomm.com/models/convnext_base
  5. github.com/pytorch/vision/blob/main/torchvision/models/c... https://github.com/pytorch/vision/blob/main/torchvision/models/convnext.py
  6. github.com/openvinotoolkit/open_model_zoo/tree/master/mo... https://github.com/openvinotoolkit/open_model_zoo/tree/master/models/public/convnext-tiny
  7. catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/pretrained_convn... https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/pretrained_convnextv2

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "ConvNeXt", https://aurion.semai.id/models/convnext-family/, records of October 2026.