README.md

metadata

title: InternVL2 Visual Extraction Demo
emoji: 🖼️
colorFrom: blue
colorTo: purple
sdk: docker
sdk_version: 3.25.0
app_file: app.py
pinned: false

InternVL2 Visual Feature Extraction Demo

This Space demonstrates the use of InternViT-6B for visual feature extraction as per the OpenGVLab/InternVL GitHub repository.

Features

• Loads InternViT-6B model for visual feature extraction
• Extracts features from uploaded or sample images
• Shows feature dimensions and processing time
• Uses Gradio for the user interface

Usage

Upload an image or use the sample image to see the visual features that would be extracted for a multimodal model.

System Requirements

• GPU: A100 recommended
• CUDA support

---

Created by the OpenGVLab/InternVL team. Implementation by mknolan.