Is WebGPU Mostly Available Now?
Check WebGPU support in Chrome, Safari, Edge, and Firefox, understand platform limits, and learn what browser support means for local OCR.
Browser-local processing / Free to use
Turn JPG, PNG, WebP, GIF, BMP, AVIF, or TIFF screenshots and scans into editable text on your device. The image never leaves your browser.
Choose an image or drag and drop
JPG, PNG, WebP, GIF, BMP, AVIF or TIFF • Up to 10 MB • 8192 × 8192 px • First animation frame, TIFF page, or AVIF image
Checks English words on your device. The selected dictionary downloads on first use. Review suggestions before replacing names or technical terms.
Workflow
Choose a clear screenshot, scan, or photo with upright, readable text. Crop it to remove unwanted content.
OCR runs on your device. Edit the extracted text, choose English (US) or English (UK), and select Check spelling. Click a word to see suggestions, replace one occurrence, or ignore it for the current language during this visit.
Review names, numbers, and technical terms against the source image. Copy the corrected text or download it as a .txt file.
Documentation
Check WebGPU support in Chrome, Safari, Edge, and Firefox, understand platform limits, and learn what browser support means for local OCR.
See how browser-based OCR uses text detection, recognition models, Web Workers, WebGPU, and WebAssembly to extract text on your device.
Learn two practical ways to extract text from a two-column document image while preserving the correct reading order.
Understand PP-OCRv6, the difference between text detection and recognition, and how lightweight AI models power browser-local OCR.
Photograph or scan a paper page, run browser-local OCR, check wording and paragraphs, and save a corrected text copy.
Learn how to extract editable plain text from screenshots and document photos locally in your web browser with data privacy.
Explore
Recognize and edit table grids on your device without uploading the selected image.
Turn document images into editable Markdown with a source editor and rendered preview.
Rotate and crop an image precisely, then download the selected region as a PNG.
Help
Use JPG, PNG, WebP, GIF, BMP, AVIF, or TIFF images up to 10 MB and 8192 × 8192 pixels. Only the first animation frame, TIFF page, or AVIF image is used. OCR extracts plain text; use Local Image to Table when you need editable rows and columns.
OCR and English spelling checks run on your device; your image and recognized text are not sent to our application API. OCR prefers WebGPU and automatically falls back to CPU when unavailable. The first OCR run downloads recognition models, and the first spelling check downloads the selected dictionary. These resources are cached for reuse. After completing a spelling check once, you can keep checking offline with that dictionary while the page remains open.
It flags English words missing from the selected US or UK dictionary. It does not check grammar, spelling in other languages, or words that are spelled correctly but used in the wrong context. Review names and technical terms before accepting suggestions.