AIbase
EN
Home
Vit Huge Patch14 Clip 224.metaclip Altogether
V

Vit Huge Patch14 Clip 224.metaclip Altogether

Developed by timm
CLIP model based on ViT-Huge architecture, supporting zero-shot image classification tasks
Image Classification
Safetensors
#Zero-shot image classification#Multimodal pre-training#Large-scale vision models
Downloads 171
Release Time : 12/23/2024

Model Overview

This model is a dual-purpose vision-language model from OpenCLIP and timm, based on the ViT-Huge architecture, trained with the MetaCLIP dataset, and supports zero-shot image classification tasks.

Model Features

Dual-framework compatibility
Supports both OpenCLIP and timm frameworks
Zero-shot capability
Performs image classification tasks without specific training
Large-scale pre-training
Trained with the MetaCLIP dataset, possessing broad visual concept understanding

Model Capabilities

Zero-shot image classification
Image-text matching
Cross-modal understanding

Use Cases

Content understanding
Automatic image tagging
Generates descriptive labels for unlabeled images
Can recognize thousands of common objects and scenes
Visual search
Text-based image retrieval
Finds relevant images using natural language queries
Achieves cross-modal retrieval without training
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
EnglishįŽ€äŊ“中文įšéĢ”中文ãĢãģんご
Š 2025AIbase