Skip to content
Search

AI workspace · Runs on your device

AI Image & Document Studio

Edit and convert images, extract text and scan pages to PDF, on your device.

Everything here runs in your browser. Files are never uploaded; OCR and background removal download their models on first use (about 4 MB and 26 MB).

or drop a file here · PNG, JPG, WebP, BMP, GIF · up to 25 MB

How to use it

  1. Choose an image, photo of a document or PDF.
  2. Crop, rotate and adjust it, remove a portrait background, or extract its text with OCR and correct it.
  3. Download the edited image (PNG, JPEG or WebP), the text, or a multi-page scanned PDF.

Good for

  • Straightening and cropping a photographed receipt
  • Turning several page photos into one PDF
  • Converting a PNG screenshot to a smaller JPEG or WebP
  • Cutting out a headshot for a CV

Privacy and limits

Images and PDFs are processed in this browser tab and never uploaded. Only the model files are downloaded, on first use.

  • Images up to 25 MB and 8,000 px on the longest side; background removal up to 12 MB and 4,096 px
  • Background removal is designed for people; other subjects often come out poorly
  • OCR is English and works best on sharp printed text; handwriting is unreliable
  • Upscaling uses standard smoothing, not AI super-resolution

Engine: Canvas image processing; MODNet portrait matting (Apache-2.0) via Transformers.js; Tesseract OCR (LSTM) via tesseract.js; pdf.js text extraction. Results can contain mistakes. Check anything important.