timm
/

vit_huge_patch14_clip_224.metaclip_altogether

Zero-Shot Image Classification

Model card Files Files and versions

Model card for vit_huge_patch14_clip_224.metaclip_altogether

Model Details

dataset: See 'Altogether' details in https://arxiv.org/abs/2410.17251

Model Usage

This model is a dual use open_clip and timm model. Model name in OpenCLIP is ViT-H-14, pretrained_tag=metaclip_altogether.

Downloads last month: 48

Spaces using timm/vit_huge_patch14_clip_224.metaclip_altogether 3

Collection including timm/vit_huge_patch14_clip_224.metaclip_altogether

MetaCLIP

MetaCLIP & MetaCLIP2 OpenCLIP and timm models. All models are dual timm + OpenCLIP (or just timm for specific vit encoders). • 24 items • Updated Sep 19, 2025 • 3

Paper for timm/vit_huge_patch14_clip_224.metaclip_altogether

Altogether: Image Captioning via Re-aligning Alt-text

Paper • 2410.17251 • Published Oct 22, 2024