← Back to library

Florence-2 — Vision task demo

A vision demo explains its model download before a task selector and side-by-side input and output regions.

Florence-2 — Vision task demo — captured interface
Viewport capture · opening screen
Public demo · 2026-09-06Open full image ↗
Why this pattern works

Expose the download step before starting a browser model.

Design notes
Surface
A vision demo explains its model download before a task selector and side-by-side input and output regions.
Navigation
Task selector and model-loading action
Layout
Introduction above paired input and output panels
Density
Focused controls with comfortable spacing.
Color
White, slate and blue
Type
Clear heading hierarchy and readable form labels.
Pattern
Expose the download step before starting a browser model.
States
Demo before loading the model; no image supplied.
Don't
Adapt the layout to your own product and supported capabilities; write your own examples and keep important actions clearly labeled.

Make it your own

Turn this reference into a useful brief for your project. Copy it into the AI tool you already use.

Add constraints
Preview & edit prompt
Assembled in your browser. No API key or AI request needed. Changing the fields refreshes the preview.