A PDF editor that edits the text that's already there — plus signatures, stamps, redaction and built-in OCR. Windows and Linux.
- Fix the text itself. Click a line in an existing PDF and type — no white boxes pasted over the old words. Bold, italic, underline, strikethrough and color. Save, reopen, and it's still editable.
- Sign and stamp in seconds. Drop in a signature or company seal image, move it anywhere, and move it again later. Paste images straight from the clipboard.
- Scanned paper becomes searchable. A neural OCR engine is built in (Korean + English, Chinese + Japanese + English, Western European). No Tesseract, nothing extra to install, under a second per page.
- Scan straight into the editor. Works with your scanner, including multi-page auto-feed (ADF), 300–4800 DPI.
- Hide what must not leak. Redaction removes sensitive content for good; blur softens an area.
- Built for AI. Save as AI7 — a document format that keeps the page, its text, its tables (as CSV) and its structure (as Markdown) together, ready for search and AI tools.
- Edit existing text in line (L-Edit) or field (Edit) mode, with fonts sized to match the original
- Type onto scanned pages and on top of images
- Shapes: rectangle, ellipse, line, arrow — saved as vectors
- Table cells editable one by one
- Crop (non-destructive in PDF), rotate, copy and drag pages between two PDFIyagi windows
- Signature and seal images, re-editable after saving
- Redaction and blur
- Scanner support with ADF multi-page scanning
- Built-in neural OCR (PP-OCRv5) — searchable PDF output
- Page image + text layer + edit layer in one file; tables to CSV, structure to Markdown
- See AI7 format and AI7 quick start
 |
 |
| Your system | File to pick |
|---|---|
| Windows 10 / 11 | Microsoft Store |
| Ubuntu 24.04 · Debian | .deb marked ubuntu24.04 |
| Ubuntu 26.04 | .deb marked ubuntu26.04 |
| Fedora · openSUSE | .rpm |
| Arch · Manjaro | .pkg.tar.zst |
| Any other Linux | .AppImage (run without installing) or .zip |
sudo apt install ./pdfiyagi_*_amd64.deb # Ubuntu / Debian
sudo dnf install ./pdfiyagi-*.rpm # Fedora
sudo pacman -U pdfiyagi-*.pkg.tar.zst # ArchEverything the app needs, OCR models included, is bundled.