AgentDuetUse casesTemplatesPricingFAQAdd to Chrome — try free
Page tools, images & voice

Any page, understood in seconds.

Summarize, explain, translate, pull key facts or rewrite — for the page you’re on or the text you selected. Or press the microphone and just say it.

One-click actions

“For this page” offers Summarize, Explain, Translate, Key facts and Rewrite. Add your own actions for things you do often.

Ask about the page

Ask anything about what you’re reading: terms, numbers, what changed. The page switch above the message box controls whether the page is included.

Photos

Attach up to three photos to a message — with the photo button, by pasting or by dragging them into the message box — and ask about them: what’s on a screenshot, a label, a receipt, an error message. Photos are resized and stripped of metadata in your browser before sending, and aren’t stored on our servers.

Create images

Switch the message box to Image, describe what you want and choose a shape. Add a photo to use it as a reference — change its style, background or details. The picture appears in the chat; download it in full size or draw it again.

Voice

Press the microphone, speak, and check the text before sending. Clips are up to 60 seconds and aren’t stored.

Full screen

Need more room? Open the assistant in a full tab for long chats and research, then go back to the side panel with one click.

Try it on a real task

Questions

Can I ask about a photo?

Yes. Attach up to 3 photos (JPEG, PNG, WebP or GIF) in Chat or Duet mode and ask your question. The agent mode works with pages, not photos.

Can it read PDFs?

If Chrome shows the PDF as text in a tab, usually yes.

Which languages can it translate?

Most major languages.

Hand off your next errand.

Try it with a free request. Pro unlocks unlimited agent tasks, Duet and research.

Add to Chrome — try free