ConvNeXt classification architecture
The captured configuration identifies ConvNextForImageClassification with model type convnext and Transformers support.
Open Source Model Profile · facebook
convnext-small-224
convnext-small-224 is a 50.22M-parameter ConvNeXt image-classification model from facebook. According to the model card, it targets ImageNet-1k labels at 224x224 resolution.
- Publisher
- Task
- image-classification
- Model type
- convnext
- License
- apache-2.0
- Library
- transformers
- Publication status
- Approved for indexing
Model overview
convnext-small-224 is published by facebook as an image-classification model. The captured configuration identifies ConvNextForImageClassification, and Safetensors metadata reports about 50.22M parameters. According to the model card, it is a ConvNeXt model trained on ImageNet-1k at 224x224.
Recorded capabilities
Modernized ConvNet design
According to the model card, the design starts from ResNet and takes Swin Transformer inspiration while remaining a pure convolutional model.
ImageNet-1k 224 resolution
According to the model card, this model was trained on ImageNet-1k at 224x224 and predicts 1,000 ImageNet classes.
Vision tagging
Hub tags record transformers, pytorch, safetensors, convnext, vision, ImageNet-1k dataset linkage, and endpoints compatibility.
Use cases in the source record
- Raw image-classification work assigning images to 1,000 ImageNet classes.
- Fine-tune discovery for task-specific versions referenced through the model hub.
Limitations and unknowns
- No evaluation results were extracted from this record.
- According to the disclaimer in the card, the card was written by the Hugging Face team rather than the releasing team.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: facebook/convnext-small-224
Captured: Unknown. Processed: 2026-09-07T19:34:44.181813+00:00.
ConvNeXT (large-sized model) ConvNeXT model trained on ImageNet-1k at resolution 224x224. It was introduced in the paper A ConvNet for the 2020s by Liu et al. and first released in this repository . Disclaimer: The team releasing ConvNeXT did not write a model card for this model so this model card has been written by the Hugging Face team. Model description ConvNeXT is a pure convolutional model (ConvNet), inspired by the design of Vision Transformers, that claims to outperform them. The authors started from a ResNet and "modernized" its design by taking the Swin Transformer as inspiration. Intended uses & limitations You can use t…
F001F002F003F004F005F006F007F008F009F010F011F012F013F014F015