Snipping Tool vs PowerToys Text Extractor vs browser OCR

Snipping Tool ships with Windows and can pull text out of a snip. PowerToys Text Extractor is a free add-on from Microsoft that copies text from any part of the screen. Both shine when you want text from whatever is on screen right now without opening a browser tab.

Screenshot2Text runs OCR in your browser tab. That fits saved screenshots, phone transfers, scanned PDFs, and the same steps on Windows, Mac, or Linux.

No single path is "better" in every case. Pick the one that matches the file you already have and the machine you are on.

Every comparison guide on this site uses the same test set: the 22 base screenshots from our OCR accuracy test. There are 5 app and settings screens (ui01-settings to ui05-form), 5 blocks of paragraph text (para01-article to para05-policy), 4 code and terminal snippets (code01-js to code04-json), 4 receipts, invoices, and price lists (num01-receipt to num04-bank), and 4 chat conversations (chat01-sms to chat04-family).

The table shows mean character error rate (CER) for each kind of screenshot saved as a lossless PNG, as a dark mode PNG, and as a simulated phone photo of the screen. Lower is better, and 0.00% is a perfect read. The main figures are run 3 results from September 25, 2026, read by the same in-browser OCR the homepage tool uses after a second OCR update that day. Run 2 results from earlier the same day sit next to them as history. We calculated the per-kind averages from the run 3 CSV and the run 2 CSV. The run 1 baseline CSV is there too.

In run 3 on this set, plain paragraphs stayed close to perfect in every version, from 0.09% as a PNG to 0.87% as a phone photo, and the 4 chat screenshots read with no errors as PNG. Dark mode raised the error on app screens from 0.41% to 3.87%. Simulated phone photos hurt most, above all chat screenshots at 21.03% and app screens at 12.87%. Across all 22 screenshots, run 3 came in at 0.65% as PNG, 1.77% in dark mode, and 9.74% as phone photos. In run 2 those three were 1.16%, 2.22%, and 10.12%.

Know the limits before you lean on these numbers. We generated the screenshots in Chromium with Linux fonts, they have no phone status bars, and the phone photos are simulated rather than taken with a camera. With 4 or 5 screenshots per kind, one bad read moves an average.

We did not run Snipping Tool or PowerToys Text Extractor on these screenshots. Both are Windows apps and our test runs on a Linux machine, so there are no Windows numbers in this table and no head-to-head result. The table describes Screenshot2Text only.

Text Extractor is one of the utilities in Microsoft PowerToys, a free, open-source set of Windows tools made by Microsoft. It is not part of Windows itself. You install PowerToys from the Microsoft Store, from its GitHub releases page, or with WinGet. Per Microsoft's PowerToys install page, it runs on Windows 11 and on Windows 10 version 2004 or newer, on a 64-bit x64 or ARM64 PC.

To turn it on, open PowerToys settings and make sure Text Extractor is enabled. Then press Win+Shift+T, the default shortcut (you can change it in settings). An overlay covers the screen. Click and drag a box around the words. When you let go, the text in that box goes to your clipboard and the overlay closes. Press Esc to back out.

It reads text with the Windows OCR engine, so it only recognizes languages that have a Windows OCR language pack installed. By default it follows your Windows language, and you can pick a preferred language in its settings. Adding a new pack is a PowerShell step that needs an administrator prompt.

Microsoft's Text Extractor documentation says the output may not be perfect and needs a quick proofread. It also points you to Snipping Tool if what you want is a saved screenshot rather than just the text.

Snipping Tool takes a capture first. Press Win+Shift+S, snip the area, then select Text actions in the Snipping Tool window. You can select part of the text or use Copy all text. Quick redact can hide email addresses and phone numbers before you share the image. Microsoft's Snipping Tool help says all text recognition runs locally on your device. Text Actions arrived in the Windows 11 version of Snipping Tool, as the Windows Insider announcement describes, so you need Windows 11 and an up-to-date Snipping Tool. Microsoft's guide to Snipping Tool notes that language support for text detection may be limited.

Text Extractor skips the capture window. You drag a box and the words land on the clipboard in one step. That is quicker for a line or two, but you never get a picture to check or redact.

The catch is the install. Snipping Tool is already there on Windows 11. PowerToys is a separate app, so a work PC where IT controls installs may not allow it.

Both Windows tools start from what is on the screen. A browser tool starts from a file or a paste, works on any OS or a locked-down machine with nothing to install, reads PDFs, and can run several files in one batch.

This table compares features only. It is not an accuracy test. Our own measured numbers are in the 22-screenshot section above, and they cover Screenshot2Text only. The Windows facts come from the Microsoft pages linked in the sections above.

Use the built-in Windows path when the text is already visible on this PC and you only need a quick copy. Capture a region, run the OS text extract action, paste into your notes. No file save. No browser warm-up.

PowerToys Text Extractor suits a quick grab from a dialog, a settings pane, or a label on a busy desktop, because you drag a box right on the screen. We have not measured how accurately it reads those. If your IT image already includes PowerToys, it is a short path: one shortcut and the text is on your clipboard.

Stay there when you do not need language picks beyond what Windows offers, you are not processing a PDF stack, and you do not care about an editable box with download .txt on a phone later.

Open Screenshot2Text when the capture already lives as a PNG, JPG, WebP, or PDF. Drop the file. Or paste from the clipboard with Ctrl+V. Edit the result in the page, then copy or download text.

Browser OCR helps when you move between machines. A Chromebook, a borrowed laptop, or a Mac at home will not have your PowerToys setup. The same homepage works everywhere recognition can load.

It also helps for batches and scanned PDFs. Snipping Tool and Text Extractor are region-first. They are not a multi-page PDF reader. If you need several pages or several screenshots labeled in one pass, use the browser tool or the PDF landing.

Privacy-minded workflows that must avoid upload converters also land here. Recognition runs on-device in the tab. Ads stay on the homepage and stay separate from recognition.

Input: Snipping Tool and Text Extractor start from the live screen. Screenshot2Text starts from a file, a paste, or a PDF you already have.

Languages: the Windows tools lean on Windows OCR language support, and Text Extractor needs the matching language pack installed. The browser tool has a language picker for the packs listed on the page. Match the language to the screenshot before you judge accuracy.

Output: all three can land text on your clipboard. The browser page also keeps an editable box and a .txt download, which helps when you want to clean OCR mistakes before paste.

Scope: the OS tools fit instant desktop grabs. Browser OCR fits saved assets, phone screenshots transferred over, and PDF pages that need a consistent flow. These are workflow differences, not accuracy results.

If the words are on this Windows desktop right now and you only need them once, use Snipping Tool or Text Extractor.

If the words live in a file, a chat image, a PDF, or you want the same steps on another OS, open Screenshot2Text.

If Windows extract looks fine, keep it. Do not force a browser round trip for pride. If Windows extract misses dark UI, tiny type, or a multi-column scan, try the browser tool, and fix the capture with the cleaner-screenshot guides either way. Dark mode raised our own error rate on app screens too: from 0.41% to 3.87% in run 3 of our test on September 25, 2026, and from 0.56% to 4.03% in run 2 earlier that day.

None of these is courtroom-perfect OCR. Handwriting, neon-on-neon UI, severe blur, and curved labels still fail. Cropping and contrast matter more than which brand logo sits on the window.

Screenshot2Text is free to use. Windows tools have their own OS limits and update cadence. Compare features, not slogans.

If recognition fails in the browser after a sharp crop and the right language, read the failure checklist guide before you assume the site is broken.