Chrome on Mac: Google Lens or on-device text capture?

If you browse in Chrome, you already have a text reader built in: Google Lens. It can read the words in any part of a tab, and it does far more than that. It also works on Google’s servers, which is fine for a recipe photo and less fine for an internal dashboard. Here’s how Lens works in Chrome on a Mac, when on-device recognition is the better fit, and how to handle the places in Chrome where text is hardest to reach.

Chrome’s Google Lens

According to Google’s Chrome Help, as of October 2026, there are several ways in:

• Right-click the page and choose Search this tab with Google Lens, then drag over the area with the text.
• Right-click an image and choose Search with Google Lens.
• Use the Chrome menu. Click the three-dot More menu and choose Search this tab with Google Lens.
• Click the address bar and choose Ask Google about this page, where it’s offered.

Results open in a side panel. Google’s help page describes Lens as a way to search what you select, and it says Google must be your default search engine for Lens to work. Depending on your Chrome version, the Lens view may also let you select the text it found to copy or translate it, but that part of the interface changes from time to time.

Lens is the better pick when you want answers rather than the words themselves: finding a product in a photo, identifying a plant, or searching for where an image came from. Our comparison of Live Text and Google Lens goes through those differences.

When on-device matters

Google’s help page is clear about the data: when you search a whole tab with Lens, a screenshot of the page and page data are sent to Google. Google says it keeps that only temporarily and doesn’t use it for training. For a public web page that’s a small trade. For a bank portal, a customer record or a company dashboard, you may not want a screenshot of it leaving your Mac at all.

In Unpixel

Unpixel reads text on your Mac with Apple’s Vision framework. Press ⇧⌘2, drag over the text in the Chrome tab and let go; the words are on your clipboard. There’s no search engine requirement, no side panel, and no account, and the capture itself never leaves your Mac. It also works outside Chrome, in a call, a PDF reader or a game, with the same shortcut.

What it won’t do is search the web for you or tell you what a picture shows. If that’s the job, Lens is the right tool.

Images on a page

Live Text belongs to Apple’s own apps rather than Chrome, but Chrome gives you two easy ways to get an image somewhere that can read it:

• Copy the image. Control-click it and choose Copy Image. In Preview, choose File → New from Clipboard and use Live Text to select the words.
• Read it straight from the clipboard. After Copy Image, Unpixel’s From Clipboard reads the copied picture directly. That helps with a large infographic that’s shown shrunk on the page, because the copy is the full image rather than what fits in the window.

PDFs in Chrome

Chrome opens PDFs in its own viewer. If the PDF has real text, select it and copy as usual. If dragging selects nothing, the PDF is probably a scan, which is a stack of page images.

For a scan, download it first. Preview can then make it searchable with File → Export and Embed Text, or you can open it with Unpixel’s Open Image…, which reads up to 50 pages per capture and puts all their text on the clipboard. Our page on Preview compares the two.

Web apps and canvas

Many web apps in Chrome draw their content into a canvas: whiteboards, design tools, some charts and dashboards. To Chrome, that text is a drawing, so it can’t be selected. Chrome’s reading mode (right-click a page and choose Open in reading mode, or More → More tools → Reading mode) helps with articles whose layout makes selecting awkward, but it can’t turn a canvas drawing into text.

For canvas text, capture it from the screen. If the app shows a table, such as a report or an admin list, select the whole grid with ⇧⌘2: Unpixel recognizes tables and they paste into Google Sheets, Numbers or Excel as cells. For tables that are real HTML, copying them directly is better; see copying a table from a website.

Videos and captions

Live Text on paused web video is a Safari feature, not a Chrome one. In Chrome, pause on a clear frame and capture the text instead; the YouTube player has its own tricks for getting a sharp frame, covered on our YouTube page.

For spoken words, Chrome has Live Caption. Open the More menu, choose Settings, then Accessibility, and turn on Live Caption. Google says the audio and captions are processed on your device, never stored, and never leave it. Streaming services that protect their video may appear black to screenshots and capture tools in Chrome too; for those, use the service’s own subtitles.