ImageNet-22K 384-resolution tuning
According to the model card, the model was pretrained with the FCMAE framework and fine-tuned on ImageNet-22K at 384x384 resolution.
Open Source Model Profile · facebook
convnextv2-base-22k-384
convnextv2-base-22k-384 is a ConvNeXt V2 image-classification model from facebook. According to the model card, it was pretrained with FCMAE and fine-tuned on ImageNet-22K at 384x384.
- Publisher
- Task
- image-classification
- Model type
- convnextv2
- License
- apache-2.0
- Library
- transformers
- Publication status
- Approved for indexing
Model overview
convnextv2-base-22k-384 is published by facebook as an image-classification model. The captured configuration identifies ConvNextV2ForImageClassification with model type convnextv2, and Safetensors metadata reports about 88.7M parameters under apache-2.0. According to the model card, it was pretrained with FCMAE and fine-tuned on ImageNet-22K at 384x384.
Recorded capabilities
FCMAE and GRN design
According to the model card, ConvNeXt V2 introduces a fully convolutional masked autoencoder framework and a Global Response Normalization layer.
Documented classification use
According to the model card, the raw model can be used for image classification, including the documented COCO-image example mapped to 1,000 ImageNet classes.
Use cases in the source record
- Raw image-classification workflows using the documented Transformers image-processor path.
- Fine-tune discovery for task-specific versions, as suggested by the card's pointer to the model hub.
Limitations and unknowns
- No context-window value was extracted from this record.
- No evaluation results were extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
- The card notes it was written by the Hugging Face team because the releasing team did not provide one; design claims were not independently verified.
Source and provenance
Source: facebook/convnextv2-base-22k-384
Captured: Unknown. Processed: 2026-09-07T19:34:44.243715+00:00.
ConvNeXt V2 (base-sized model) ConvNeXt V2 model pretrained using the FCMAE framework and fine-tuned on the ImageNet-22K dataset at resolution 384x384. It was introduced in the paper ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders by Woo et al. and first released in this repository . Disclaimer: The team releasing ConvNeXT V2 did not write a model card for this model so this model card has been written by the Hugging Face team. Model description ConvNeXt V2 is a pure convolutional model (ConvNet) that introduces a fully convolutional masked autoencoder framework (FCMAE) and a new Global Response Normalization…
F001F002F003F004F005F006F007F009F010F011F012F013F014F015F016