AIbase
EN
Home
Openvision Vit Huge Patch14 84
O

Openvision Vit Huge Patch14 84

Developed by UCSC-VLAA
OpenVision is a fully open, cost-effective family of advanced vision encoders designed for multimodal learning.
Image Classification
Transformers
Open Source License:Apache-2.0#Multimodal Vision Encoding#Cost-Effective Models#Open Weights
Downloads 19
Release Time : 5/6/2025

Model Overview

OpenVision provides a series of efficient vision encoders suitable for multimodal learning tasks, aiming to achieve high-performance visual feature extraction at low cost.

Model Features

Fully Open
The model is completely open, allowing free use and modification.
Cost-Effective
Achieves high-performance visual feature extraction at low cost.
Multimodal Support
Designed for multimodal learning tasks, supporting integration with other modalities.

Model Capabilities

Image Feature Extraction
Multimodal Learning

Use Cases

Computer Vision
Image Classification
Use OpenVision to extract image features for classification tasks.
Multimodal Learning
Combine text and image data for multimodal learning.
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase