Skip to content

Chrome Local AI

Color mode

Image analyzer

Ask the model what is in a picture.

Prompt API …
Experimental: it may not be available in your Chrome.

What it does

It uses the image input of the Prompt API (experimental): the model looks at your picture and answers a question about it. The image is processed on your device and never uploaded.

Good to know: Image input depends on the Chrome version and the model. If you get an error, the model on your device probably cannot read images yet.

How to use it

  1. Choose an image file. A preview appears.
  2. Write a question, or pick one of the suggestions.
  3. Press Analyze. The answer streams in.

Answer

The image is read by the model on your device; it is never uploaded.