Pluck — On-device OCR

Pluck the text out of any image on a page, using Chrome's built-in Gemini Nano. Nothing leaves your device.

Someone sends you a screenshot when what you needed was the text. Pluck reads it for you. Click the Pluck icon and every image on the page gets a small OCR badge. Click a badge, and the text in that image streams into a panel, ready to copy. Addresses, booking references, error messages, slides, signs, receipts: if you can read it, Pluck can copy it. PRIVATE BY DESIGN Pluck uses Gemini Nano, the AI model built into Chrome, which runs entirely on your own computer. Your images are never uploaded, there's no account, and there's no server to send anything to. HOW IT WORKS • Click the toolbar icon, or press Alt+Shift+O, to switch Pluck on for the current tab • Click the OCR badge on any image to read its text • Copy the result with one click • Images you've already read are marked with a tick, so you can click again to see the text instantly • Click the icon again to switch Pluck off SCREENSHOTS ON YOUR COMPUTER Drag the image file into a Chrome tab, then use Pluck as normal. For this to work, turn on "Allow access to file URLs" in Pluck's extension details (a one-time setting). MINIMAL PERMISSIONS Pluck only runs on a tab when you click its icon. It doesn't read your browsing, doesn't run in the background, and doesn't collect any data. REQUIREMENTS Pluck uses Chrome's built-in AI, so it needs: • A recent version of Chrome on Windows, macOS, Linux or a Chromebook Plus • About 22 GB of free disk space, and either a GPU with more than 4 GB of video memory or 16 GB of RAM • The first time you use it, Chrome downloads the Gemini Nano model (a few GB). After that, it's ready in seconds. If something isn't set up, Pluck tells you exactly what's missing and how to switch it on. Pluck is open source: https://github.com/conroyp/pluck-chrome-ocr