convnextv2-base-1k-224

Model overview

facebook publishes convnextv2-base-1k-224 as an image-classification model on the Transformers stack. Captured config identifies ConvNextV2ForImageClassification with a convnextv2 model type, and Safetensors metadata reports 88,717,800 parameters. Card data records apache-2.0. The captured model card, which notes it was written by the Hugging Face team, describes a base-sized ConvNeXt V2 checkpoint pretrained with FCMAE and fine-tuned on ImageNet-1K at 224x224.

Recorded capabilities

Use cases in the source record

• Image classification into the card's documented 1,000 ImageNet classes using AutoImageProcessor and ConvNextV2ForImageClassification.

Limitations and unknowns

• No evaluation scores were extracted from this record.
• Provider state is historical snapshot data, not independently refreshed current availability.
• The captured model card states it was written by the Hugging Face team rather than the original ConvNeXt V2 authors. Pretraining, fine-tuning, architecture narrative, and benchmark-improvement language are card claims, not independently verified Ethen facts.

Source and provenance

Source: facebook/convnextv2-base-1k-224

Captured: Unknown. Processed: 2026-09-07T19:34:44.216715+00:00.

ConvNeXt V2 (base-sized model) ConvNeXt V2 model pretrained using the FCMAE framework and fine-tuned on the ImageNet-1K dataset at resolution 224x224. 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…