Skip to content
Image To Text

Free browser-based OCR — nothing leaves your device

Get text from an image as editable text

Pull text from photos, screenshots and scanned documents and turn it into plain text you can copy, edit and search. Drop an image, choose the language, and get your text in seconds — no account, no upload, everything runs locally.

Supports 30 languages including English, Spanish, Chinese, Arabic and Japanese. Process up to 20 images in a single batch.

  • 100% free
  • No signup needed
  • 30 languages
  • Up to 20 images
  • Runs offline
Line breaks

How to get text from an image

Three steps to pull text from any image and turn it into editable, searchable text — no account needed, no software to install.

  1. Load your image

    Drag a photo, screenshot or scan onto the drop zone, click to browse your files, or paste from the clipboard with ⌘V / Ctrl+V. You can add up to 20 images at once.

  2. Choose the text language

    Select the language of the text in the image — English is the default, with 30 languages available. Matching the language improves accuracy, especially for non-Latin scripts.

  3. Copy or save the result

    The recognized text appears in an editable box with a confidence score. Hit Copy text to grab it, Download .txt for a file, or Download all when you have multiple images.

Technical specifications

Accepted image formatsJPG/JPEG, PNG, WebP, GIF, BMP, AVIF
Maximum file size25 MB per image
Output formatEditable plain text — copy to clipboard or download as .txt
Supported languages30, from English and Spanish to Arabic, Hindi and Japanese
Batch limit20 images per batch
OCR engineTesseract 5 (LSTM) compiled to WebAssembly
Processing locationYour browser — no image leaves your device
PriceFree, no signup, no watermark, no usage cap
Works offlineYes, after the engine and language pack have loaded once

Frequently asked questions

How do I get text from an image?
Drop or paste the image on this page, pick the language, and the text is recognized in seconds. You can then copy it to your clipboard or download it as a plain-text file.
Can I extract text from a photo taken with my phone?
Yes. Phone photos, screenshots, scanned pages and any standard image file work. For screenshots you can paste directly with ⌘V or Ctrl+V — no need to save a file first.
What image formats are accepted?
JPG/JPEG, PNG, WebP, GIF, BMP and AVIF, up to 25 MB each. This covers virtually every format from phones, cameras, scanners and screenshot tools.
Is there a file-size limit?
Each image can be up to 25 MB. Because everything runs locally in your browser, there is no server-side upload cap or bandwidth restriction.
How accurate is the text extraction?
Clean printed text at decent resolution typically scores above 90% confidence. Accuracy drops with blur, unusual fonts, tight line spacing or cluttered backgrounds. Each result shows its confidence score so you know where to double-check.
Are my images uploaded anywhere?
No. The OCR engine runs entirely in your browser using WebAssembly — no image data is transmitted, stored or logged. Closing the tab wipes everything from memory.
Can I process several images in one go?
Yes. Add up to 20 images in a single batch. They are processed one after another, and you can copy or download all results together.
Which languages are supported?
30 languages, including English, Spanish, French, German, Portuguese, Russian, Arabic, Hindi, Chinese (Simplified and Traditional), Japanese, Korean, Thai, Vietnamese, Ukrainian, Turkish and Hebrew.

How text extraction from images works

When you load an image, the browser decodes it, scales it to the resolution range the OCR model expects, and passes it to Tesseract 5 — the established open-source recognition engine, compiled to WebAssembly so it runs at near-native speed inside your browser tab.

The engine identifies lines, words and characters, then returns plain text together with a confidence score for each image. Results appear in editable boxes so you can fix any misread character before copying or downloading.

The first run takes a few extra seconds while the engine (~3 MB) and the chosen language pack (0.5–4 MB) download. Both are cached by your browser, so every subsequent image processes immediately.

Looking for the main tool? Image to Text is the same engine with the same features, or try extract text from image and text in image to text for alternative starting points.

Your images stay private

All text recognition runs entirely on your device. No image is transmitted, stored or logged — the only downloads are the OCR engine and the language pack you selected. Closing the tab releases all in-memory copies.