Gemini in Chrome Review 2026: Pricing, Features and Verdict

What it is

Gemini in Chrome is Google's Gemini assistant built into the Chrome browser. It opens from an Ask Gemini button at the top of the browser, from a keyboard shortcut, or from the Mac menu bar and Windows system tray once you opt in. Google is clear that it is different from visiting gemini.google.com: only the built in version can read your tabs and run Live voice on top of the page.

It has two halves:

• Chat with your tabs. Ask about the page you are on, compare pages, get takeaways, draft a Gmail message or ask about a video you just watched.
• Auto browse. Hand Gemini a multi step task and it plans, opens sites and acts for you while you watch or do something else.

To use it you need to be 13 or older (or the local age), in a supported region, on the latest Chrome on a Chromebook Plus, Mac or Windows computer, signed in to Chrome and using a supported language. Work and school accounts need an admin to turn it on.

Pricing explained

Option	Price	What you get in Chrome
Gemini in Chrome	$0	Chat with your current tab and up to 10 open tabs, Gemini Live, media understanding
Google AI Pro	$19.99 per month	Chrome auto browse, up to 20 multi step task requests per day
Google AI Ultra	$99.99 or $199.99 per month	Highest access to auto browse, up to 200 task requests per day

Auto browse requires a personal Google account with AI Pro or Ultra. It is not available with a school account, and work accounts depend on the admin. Google's plans page lists Chrome auto browse on Pro and "highest access" on Ultra.

The value math depends on what else you use. Google AI Pro bundles Gemini's Pro model, Gemini in Gmail and Docs and 5 TB of storage according to our Gemini record, so if you already pay for Google storage, auto browse comes close to free. If you only want an agent for occasional bookings, 20 tasks a day is more than most people need, and the free chat side may be all you use. Compare costs across tools on the Price Index.

Key features

Tab context

By default Gemini uses the page in your current tab. You can add up to 10 recent open tabs, mention one with @, or remove a tab from the conversation. A glowing underline marks each tab Gemini is reading. If you share a Workspace page such as a Google Doc, Gemini may read the whole document from your account rather than only what is on screen.

Auto browse

You describe the task, review the plan Gemini will use, then click Start Task. Google's examples include comparing products and adding items to a cart, finding and booking accommodation, making restaurant reservations, scheduling appointments and fetching receipts. While it works, the task tab shows an auto browse icon. If it needs you, a notification appears and you can take over and hand the task back.

Auto browse can run in your local Chrome, with access to the sites you are signed in to, or in a remote browser in the cloud that keeps going if you close your laptop. Mobile uses the remote browser only.

Sign in help

With your permission, Gemini in Chrome can use Google Password Manager and Sign in with Google to sign in to sites. Google says it does not share your passwords with Gemini or other sites, and you can review and remove the sites it can sign you in to.

Gemini Live and media

You can talk to Gemini while you browse with Live (not for users under 18). Media understanding, on by default, lets you ask about audio and video you already played on media sites.

Google app connections

Gemini can surface a flight time from Gmail, check it in Google Flights and draft an update, look up past photos and albums, and draft messages in Gmail, all without switching tabs.

What users say

Gemini in Chrome has no separate app store listing, so there is no rating that isolates it. The Gemini app it shares an account and models with holds 4.7 on the App Store from about 2.35 million ratings, per our Gemini review. Public discussion splits in two: people who like getting AI without switching browsers, and people who want a way to avoid AI in Chrome entirely. Google answers the second group with settings to hide the icon and turn permissions off.

Privacy and security

When it runs. Google says Gemini in Chrome activates only when you click the Gemini icon or use your shortcut.

What it collects. When you use it, Gemini collects page content and the URL from your current tab and any tabs you share, and relevant URLs from your history if you ask it to find pages you visited. Google notes that some page content Gemini uses might not be visible to you. Site info, audio and files are stored in Gemini Apps Activity if Keep Activity is on; page content is logged to your account temporarily and does not appear in that activity list.

What auto browse can do. In your local browser it has the same access you do, including signed in sites, and it may share your name, contact details, files and preferences with sites to finish a task. Google's own help page lists prompt injection as a risk, with examples like an instruction hidden in a page telling the agent to send your Gmail messages to an outside service or post private info publicly.

Safeguards. Gemini asks you to take over for finalizing financial transactions, accepting terms of service and creating accounts, and asks for confirmation before certain other actions. Safe Browsing must be on Standard or Enhanced for auto browse to run. Google still states plainly: "You're responsible for Gemini's actions during a task, including mistakes and unexpected results like purchases."

Controls. You can turn off "Let Gemini browse for you," stop sharing tabs, turn off sharing the current tab by default, and manage or delete Gemini Apps Activity. Parents using Family Link can turn off a teen's access.

Who it is for

• Chrome users who do not want a new browser. This is the lowest friction way to get an AI browser.
• Google AI Pro subscribers. Auto browse is already in your plan.
• Gmail, Calendar and Photos users. The Google app hooks are deeper than any rival's.

Who should skip it

• Privacy maximalists. Shared page content and URLs go to Google, and the feature is tied to a signed in Chrome profile.
• Heavy agent users on a budget. 20 tasks a day on Pro is fine for errands, not for bulk work; Manus runs 20 tasks at once on its $20 plan.
• Non English users who want auto browse. It requires English as the device language.
• Anyone who wants a model choice. Opera Neon and Comet offer models from several labs.

Alternatives

• Comet: free agentic browser on four platforms.
• Dia: published prompt injection controls and SOC 2 Type II.
• Opera Neon: model choice and MCP for $19.90 per month.
• ChatGPT agent: OpenAI's Work agent with its own browser in the ChatGPT desktop app.

See our ranking of the best AI browsers.

Verdict

Gemini in Chrome is the default choice for anyone already living in Chrome and Google's apps: free chat with real tab context, and an auto browse agent that shows its plan before it acts and hands back the risky steps. Pay for Google AI Pro only if you will use auto browse or already want the storage and Gemini perks. Keep in mind Google's own words that the agent is experimental and that you own its mistakes.