Edit PDFs.
Keep them yours.

The open-source alternative to Adobe Acrobat and iLovePDF. Every page renders and every edit saves on your machine — documents are never uploaded.

No account, no upload, no install — it runs in the tab.

Features

Every tool

This is the whole list — everything the app can do today, and nothing it can't. Each of these works offline, on the desktop app and (except where noted) in the Chrome extension.

Read & navigate

Opening a 2,000-page document doesn't render 2,000 pages.

Open & viewTiled rendering, smooth scrolling, zoom and fit.
Find in documentMatch case, whole word, a result list — and it finds phrases broken across a line wrap.
ContentsThe document's own bookmarks, with the section you're reading highlighted.
Page thumbnailsJump around, and reorganise from the same panel.
TabsSeveral documents open at once.
Undo & redoAcross every kind of edit, not just markup.

Annotate & comment

Markup snaps to the glyphs you actually selected, not to a hand-drawn box.

HighlightOver real text selection, across line wraps. PDF annotator
UnderlineSame selection model as highlighting.
StrikeoutFor marking deletions in a draft.
Sticky notesComments anchored to a spot on the page.
Freehand drawingInk, with a smoothed stroke.
RectangleOutline or filled; the drag sets the size.
EllipseInscribed in the box you drag.
EraserRemove markup you've added.
Comments panelEverything marked up, in one list.
FlattenBake markup into the page so the recipient can't move or delete it.

Change what it says

Redaction removes the text from the file. It is not a black rectangle you can select underneath.

Edit textReplace a line in place, with the substituted font called out when one is.
Move textDrag a run to a new position.
Move imagesReposition a placed image.
RedactTrue content removal — the words are gone from the bytes. How to redact a PDF

Organise pages

MergeCombine several documents into one.
Split & extractPull a page range out into its own file. How to split a PDF
ReorderMove pages up and down. Organize PDF pages
RotateAny page, any quarter turn.
Delete pagesRemove what you don't need.
CropTrim the visible area of a page.

Forms & signing

Fill formsText fields and checkboxes; radio buttons best-effort.
Add a text fieldTurn a flat document into a fillable one.
Add a checkboxSame, for tick boxes.
Draw a signatureDraw once, reuse it, drag it onto any page.
Inspect signatures claims only Shows what a digital signature claims. It does not check that it's genuine — and says so, plainly, rather than implying otherwise.

Protect & finish

Password protectAES-256 encryption, with per-recipient permissions. Password protect a PDF
Watermark SupporterTiled text or a logo, any angle, any density.
Page numbersAny corner, any font, starting from any number.
Bates numberingZero-padded, prefixed, for discovery sets.
CompressSave a smaller copy.

Read scans & compare drafts

OCR Supporter Make a scan searchable — text you can select, search and copy. In the web app the recogniser downloads once (about 3 MB, from openpdfedit.com itself) and then works offline; the desktop app has it built in and fetches nothing at all. Not in the Chrome extension — that one-time download is exactly the kind of remote code a Manifest V3 extension may not fetch.
Compare documentsText differences and rendered-pixel differences, side by side.

Take the text out

Every hosted converter wants the document uploaded first. This one reads it where it is.

Export to MarkdownHeadings, lists and tables, read from the document's own structure — for a note, a wiki, an Obsidian vault. A scan gives paragraphs instead: nothing in a picture of a page says which line was a heading.
Export to plain textThe words, in order, and nothing else.
40 tools. No sign-up, no upload, no page limit. Working on: Office conversion, and typing straight into a form field on the page.

How it works

Every step happens on the machine you are reading this on.

There is no upload, no queue and no job id, because there is no server to send a document to. Four steps, and none of them leaves the room.

1 · Open a PDF From your disk, straight into the tab or the app. The file is read where it already is.
2 · It renders locally The same Rust core either way — native on the desktop, compiled to WebAssembly in the browser.
3 · Edit it Annotate, redact, fill, sign, reorder, merge. Undo reaches back through all of it.
4 · Save it back Incrementally, so the original bytes survive and an existing digital signature still verifies.

Privacy

Built to need nothing from you but the document.

Free, and open source. You can read the code that handles your documents before you trust it with one.

Nothing is uploaded Your document never reaches a server. Every edit happens on your own machine — in the app, or in the page itself. The Chrome extension asks for no permissions at all, not one. The source is public; check.
No account, ever Not to install it, not to use it, not to save a file.
No watermarks or page caps The output is your document. Nothing of ours is stamped on it, no page limit, no export cap — a free file is a finished file.
No AI you didn't ask for There is none in it at all.

Download

Three ways to run it

One Rust core, compiled three ways — native for the desktop, WebAssembly for the browser. Same rendering, same editing, same documents-never-leave-your-machine. Pick whichever suits you; nothing is held back in any of them.

No install

In your browser

The editor as a plain web page. Your documents are never uploaded because there is nowhere to upload them to — the page does the work, not a server. Load it once and it keeps working with no network at all.

On a phone it is the same editor, laid out for the screen: tools along the bottom, pinch to zoom, and Add to Home Screen if you would rather it opened like an app.

Open the web app Chrome, Edge, Firefox, Safari · saving over the original needs Chrome or Edge
Full version

On your desktop

A native app for macOS and Windows, with the whole toolset and the fastest rendering. Free, open source, and the same build for everyone.

Coming back soon The desktop download is coming back soon. Until then, use the web app or the browser extension.
In the tab

As a browser extension

Open a PDF from a page or from disk and edit it without leaving the tab — the same editor, one click from wherever the file already is.

Add to Chrome Add to Edge Both from the browser’s own store. Firefox and Safari are not supported.
Two Supporter tools — watermark and OCR — are 1,000 credits once, together. Everything else is free. Read the source

Pricing

Free, and the two exceptions are named here.

No subscription, no account to create, no page limit, no export cap and nothing of ours stamped on your document.

Free Viewing, annotating, editing text and images, redacting, filling and signing forms, reorganising pages, merging, splitting and comparing. All of it, on every platform.
Supporter · 1,000 credits, once Watermarking and OCR, together. One payment, not a subscription — and the rest of the app never asks.

FAQ

Questions

The ones that actually get asked about an editor that doesn't upload anything.

Is OpenPdfEdit free?

Yes, apart from two tools. Annotating, editing text, redacting, filling and signing forms, reorganising pages, merging, splitting and comparing are all free, with no page limit, no export cap and nothing of ours stamped on the output. Two Supporter tools — watermarking and OCR — cost 1,000 credits once, for both together.

Are my documents uploaded anywhere?

No. Opening, annotating, redacting and saving all run on your own machine — in the app, or in the page itself. No document, and no part of one, is sent anywhere. The Chrome extension declares no permissions and no host permissions at all, so it cannot reach a server even if it wanted to. The source is public and the claim is checkable.

Do I need an account?

Not to install it, use it, or save a file. An account exists only for the two Supporter tools, and only to carry the one-time payment for them.

Does it work offline?

Yes. The desktop app never needs a network. The web app needs one only to load the page the first time; after that it keeps working with the network off, including after a reload.

Can it edit the text already in a PDF?

Yes. You can replace a line in place, move a run of text, and reposition an image. A PDF stores glyphs rather than paragraphs, so an edited line does not reflow the ones around it, and when the original font isn't embedded the substitute is named rather than swapped in quietly.

Is the redaction real, or a black box over the text?

Real. The text is removed from the page's content stream, including text inside form XObjects, and the covered pixels of an image are overwritten rather than the image being hidden. The file is then rewritten rather than appended to, so the removed content is not recoverable from an earlier revision. The box drawn on top is a second line of defence, not the mechanism.

Can it read a scanned document?

Yes, in the desktop app and the web app — OCR adds a searchable, selectable text layer under the scan without changing a pixel of how the page looks. Twelve languages ship with it, including Simplified and Traditional Chinese, Japanese, Korean, Arabic and Russian. Recognition runs on your machine like everything else. It is one of the two Supporter tools, and the only tool the Chrome extension doesn't have.

What does it run on?

A native app for macOS (Apple Silicon) and Windows; a web app in Chrome, Edge, Firefox and Safari; and a Chrome extension. On a phone the web app lays itself out for the screen and can be added to the home screen. Saving over the original file, rather than downloading a copy, needs Chrome or Edge.

How is it different from the online PDF tools?

Those upload your file to a server, do the work there, and send it back — which is why they can meter it by page and by day. This does the work where the file already is, so there is no queue, no size limit, no daily cap, and no copy of your document on someone else's disk.

Is it open source?

Yes, under AGPL-3.0-or-later, on GitLab. The same build everyone else gets — you can read it, build it yourself, or fork it. If you pass your version on to other people, its source has to be open too.