README.md

metadata

title: Nvidia Eagle2 2B VLM Demo
emoji: 🦅
colorFrom: blue
colorTo: indigo
sdk: gradio
app_file: app.py
pinned: false
license: mit
short_description: Multimodal Visual QA with NVIDIA Eagle2-2B

🦅 NVIDIA Eagle2-2B & LocateAnything VLM Demo

Interactive web demo for NVIDIA Eagle2-2B and LocateAnything-3B Vision-Language Models.

Features

• Visual QA & Bug Detection: Analyze web screenshots for layout bugs, visual defects, and UX errors.
• UI Grounding & Bounding Boxes: Detect interactive DOM elements, buttons, and form inputs visually.
• Custom Multimodal Prompts: Query the model with image + text prompts.