Available now

Florence-2 Open-Vocabulary Detection: object detection

Microsoft's Florence-2 looks for the object you name in an image and marks where it is with a labeled box.

Model ID

Input

Fields marked * are required. Review the price before generating.

The image to be processed.

Uploading a file needs an account. Create account

JPEG, PNG, WebP · up to 25 MB

Text input for the task

Example. Edit it or write your own.

Price$0.60 / run

You sign up first. Your settings are kept.

About this model

Overview

Upload an image and type what to find, in your own words, such as "red umbrella" or "coffee cup". Florence-2 returns a box with a label around each match, as a list of positions and as a copy of your image with the boxes drawn on it.

What you can build

  • Checks that a product or logo appears in a photo
  • Box positions for cropping a subject out of many images
  • Labels for building or reviewing an image dataset

Prompting tips

  • Name the object plainly and add one detail if there are look-alikes, such as "the white car".
  • Search for one kind of object per run so the labels stay easy to read.

What to check

Review every detection against the source. Check false matches and omissions; a model label does not establish that an object is present.

Inputs

  • Image (required): The URL of the image to be processed.
  • Text Input (required): Text input for the task.

Output and limits

  • Small or dense text in the image can come out misspelled.
  • Hands, crowds and complex poses can show artifacts; generate a few variations and pick the best.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • up to: $0.60 / run