Tiny ConvNeXt at 224x224
According to the model card, this is the tiny-sized ConvNeXt model trained on ImageNet-1k at 224x224 resolution.
Open Source Model Profile · facebook
convnext-tiny-224
convnext-tiny-224 is a tiny ConvNeXt image-classification model from Facebook. According to the model card, it is trained on ImageNet-1k at 224x224 resolution for 1,000-class classification.
- Publisher
- Task
- image-classification
- Model type
- convnext
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
convnext-tiny-224 is published by Facebook as an image-classification checkpoint. Captured configuration identifies ConvNextForImageClassification with model type convnext, and hub tags record ImageNet-1k association. According to the model card, it is the tiny ConvNeXt model introduced in A ConvNet for the 2020s by Liu et al.
Recorded capabilities
Modernized ConvNet design
According to the model card, it is a pure convolutional model inspired by Vision Transformer design, modernized from a ResNet with Swin Transformer inspiration.
Documented classification workflow
According to the model card, the model classifies a COCO 2017 image into one of 1,000 ImageNet classes with ConvNextImageProcessor and ConvNextForImageClassification.
Transformers stack with Apache-2.0
The hub record lists the Transformers library with an Apache-2.0 license, and tags record PyTorch and TensorFlow compatibility.
Use cases in the source record
- Raw image classification into one of 1,000 ImageNet classes, which the model card lists as the raw-model use, with a documented Transformers ConvNextImageProcessor workflow.
- Starting point for task-specific fine-tunes discovered through the model hub, which the model card points to for interested tasks.
Limitations and unknowns
- The model card carries a disclaimer that the releasing team did not write it and that it was written by the Hugging Face team, so publisher attribution should be read with that provenance.
- No evaluation results were extracted from this record.
- No parameter count, quantization, or context-window value was extracted.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: facebook/convnext-tiny-224
Captured: Unknown. Processed: 2026-09-07T19:34:44.196223+00:00.
ConvNeXT (tiny-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 th…
F001F002F003F004F005F006F007F008F009F010F011F012F013F014