Text that is actually yours
Click the page, type, and choose font, size, colour and alignment. You can also rewrite text already in the PDF: it stays vector.
Open source · No account · 100% free
Rewrite text that is already there, add images, shapes and signatures. Reorder pages, highlight, erase and download a real PDF. All on your device, with no upload and no watermark.
Upload
0 bytes
Your PDF, finally free.
YouProcessed in the browser
Everything you need
The features you actually use, built to run locally and produce a clean PDF. No subscriptions, no uploaded files, no surprises.
Click the page, type, and choose font, size, colour and alignment. You can also rewrite text already in the PDF: it stays vector.
Drag in photos, logos and signatures. Add rectangles, ellipses and lines with custom fill and border.
Draw freehand or sign with a trackpad, mouse or finger. The signature is inserted as a transparent image.
Highlight passages with multiply blending. Cover old data with whiteout and type over it.
Reorder, rotate, delete and insert blank pages. Changes apply to the final file, not just the view.
Original pages stay vector. Text, shapes and images are written as native PDF content.
Need another feature? The library is open source: propose it or build it.
Open an issuePdfCanvas is a browser-based PDF editor for Chrome, Edge, Firefox and Safari. Open a PDF, rewrite existing text, add images, shapes and signatures, then download a clean, vector PDF. No upload, no account, no watermark.
Use it for contracts, invoices, forms, reports and any PDF with a real text layer. Rendering runs with PDF.js, export with pdf-lib and original text editing with PDFium. Everything stays in the tab.
How it works
Three steps, zero waiting. The PDF opens instantly because it does not have to travel anywhere.
Privacy by design.
No server receives the file. History stays in the tab and disappears when you close it.
Drag it in or choose a file. No upload: it is read directly in the browser.
Text, images, shapes, pen, highlighter, signature and page management.
A new, vector PDF with no watermark. Ready to send.
No upload
The file never leaves the browser.
Zero waiting
No conversion queue.
Real PDF
Vector, not a screenshot.
Open source
Public code, MIT license.
pdfcanvas · MIT
The engine behind this site is a standalone TypeScript library. It renders with pdf.js, manages objects and exports with pdf-lib. No framework required.
import { PdfCanvas } from "pdfcanvas";
// self-host the worker
const editor = new PdfCanvas(el, {
workerSrc: "/pdfjs/pdf.worker.min.mjs",
});
await editor.load(file);
editor.addText(80, 120, { text: "Hello!" });
await editor.download();
FAQ
If something is missing, write on GitHub. The project is public and issues are welcome.
No. PdfCanvas reads, edits and exports the document entirely in the browser. The file is never sent to a server: you can verify it in the Network tab of your developer tools.
Yes. The site and the library are open source under the MIT license. There are no accounts, subscriptions, daily limits or watermarks.
Yes, when the PDF contains real text: PdfCanvas uses PDFium, the same engine as Chrome's PDF viewer, and writes the text back into the original object. If the font cannot encode the new characters, the text is covered and reinserted as vector. Scans need OCR.
Recent versions of Chrome, Edge, Firefox and Safari, on desktop and mobile. PdfCanvas uses standard Canvas, IndexedDB and Web Workers.
Yes. The library is called pdfcanvas and is published on npm under the MIT license. It exposes a framework-free TypeScript API.
No. For privacy we keep nothing. If you reload the page, open the PDF again: unsaved changes are lost. Download the result before leaving.
Ready when you are
Free, no account and no waiting. Open a file and see what happens.
Your files stay on your device, always.