The private PDF toolkit.
Organize your PDFs without ever uploading them.

Why do PDF tasks in your browser?

Every "free online PDF editor" makes the same trade: they take your file to their server, do the work, hand you the result — and in exchange you get shown ads, capped at a few files per day, and eventually pushed into a paid tier. Your PDF sits somewhere on their infrastructure for some retention period nobody actually reads the terms on. If that PDF is a contract, an invoice, a signed statement, a court exhibit — that trade is more expensive than the paid tier ever was.

Browsers have shipped the pieces for local PDF work since 2018 (Mozilla’s pdf.js in 2011, pdf-lib since 2019, the JS engines that make big-file manipulation feasible in a tab). Nothing in the pipeline requires an upload. Your PDF is opened in the browser, edited by JavaScript in the tab, and downloaded back to your Downloads folder. No server touches it.

You can prove it yourself. Press F12 in your browser, switch to the Network tab, clear it, and merge two PDFs with the tool above. Watch the request list — it stays empty. When you click Download, the file appears in your Downloads folder. Everything happened locally.

The "free" trick most PDF tools run

What each operation actually does

Merge PDFs

Combining files is the single most-searched PDF task, and it’s usually the most sensitive — contracts, statements, application packets. The tool reads each file’s pages, stitches them together in memory using pdf-lib, and writes a single PDF. The output preserves each page’s original vector text, images, fonts, and any embedded metadata. No re-rendering, no quality loss.

Split or extract pages

Need page 3 on its own, or the first ten pages as a separate file? Splitting is just reading the pages you want and writing them to a new PDF — same local-only operation as merge. Handy when you only want to send one page of a bank statement instead of the whole thing, or extract a signed page from a longer document to file separately.

Rotate, delete, and reorder pages

Scans come in sideways. Drafts have a blank page you want gone. A packet needs its sections rearranged. All three are small edits to the PDF’s structure — page rotation is a metadata property, deletion removes an entry from the page tree, reordering shuffles the list. No pixel data is touched, so quality is byte-perfect.

Images to PDF

Convert JPG, PNG, or WebP images into a single PDF where each image becomes one page. Useful for combining receipts for expense reports, scanning a paper document via your phone camera then wrapping it as a PDF, or turning a photo collection into something a portal will accept. The images are embedded at their original resolution; no re-compression unless you explicitly resize.

Compress / reduce file size

PDF compression works by finding embedded images, re-encoding them at lower resolution or quality, and rewriting the file. Text and vector content is already efficient — the win is always on images. This tool re-encodes JPEGs at your chosen quality (with the same math the image compressor uses), then rewrites the PDF. A 40MB photo-heavy PDF often drops to 3-5MB with no visible quality change.

When to use browser vs desktop PDF tools

Both have honest use cases. Rough rule:

Seven concrete jobs this handles

1. Merge monthly invoices for a client

Drag in each month’s invoice PDFs, drag to the right order (Jan, Feb, Mar...), click Merge. Download one PDF. Send to the client. No upload = the client’s billing details never sit on Smallpdf.

2. Extract one page of a contract for reference

Signed a 40-page master services agreement? Split it, keep only the exhibit you need to reference, delete the rest. Send the extract instead of the whole document.

3. Rotate a sideways scan

The scanner grabbed the page landscape when you wanted portrait. Rotate 90 degrees, download the fixed file. Takes about four seconds.

4. Convert phone-camera photos into a receipt PDF

Snap the receipts on your phone, transfer to your laptop, drop the images into the tool, get one PDF for expense-report attachment. Nothing uploaded to a receipt-tracking SaaS in the middle.

5. Shrink a 40MB PDF for email

Most email systems cap attachments at 25MB. A photo-heavy PDF often exceeds that. Run compression at quality 80 — the file usually comes in under 10MB and looks identical at reading distance.

6. Redact sensitive data before sharing

Delete the pages that contain SSNs, account numbers, or PII you don’t want in the file’s recipient chain. Save the redacted version. This is not a full redaction tool (professional redaction removes searchable text, not just visual coverage) — but page-level removal handles most everyday cases.

7. Reorder a proposal deck exported as PDF

Your slide deck exported PDF pages in a weird order because you have hidden slides. Drag pages into the right sequence, re-download. Faster than re-exporting from Keynote or PowerPoint.

Common problems and their fixes

• "File is too big" or the tab freezes — browsers hit memory limits around 200-300MB for a single PDF. If yours is bigger, use a desktop tool or split the file first.
• Merged PDF is huge — one of the inputs was likely image-heavy at full resolution. Run compression after merging.
• Pages come out blank — usually a corrupt input PDF (some scanners produce non-standard PDFs). Try re-saving the input via Chrome’s print-to-PDF first.
• Rotate button rotates the wrong page — check which page you have selected; rotation applies only to the highlighted page(s).
• Text goes garbled after edit — vanishingly rare with pdf-lib, but if it happens, the source PDF is likely using unusual font embedding. Report the file and we’ll investigate.
• Firefox / Safari acts differently — the tool works in all modern browsers but Chrome and Edge are best-tested. If a specific operation breaks in Firefox or Safari, try Chrome first to confirm it’s a browser-specific quirk.

How to prove nothing uploads

1. Open the tool page in your browser.
2. Press F12 to open DevTools.
3. Click the Network tab, then the ⛞ (clear) button.
4. Drop a PDF into the tool.
5. Run any operation (merge, split, rotate).
6. Click Download.
7. Look at the Network panel — no POST or PUT requests appear. The download itself is a local blob URL (starts with blob:), not a network fetch.

Try the same test on Smallpdf or iLovePDF and you’ll see the upload happen the moment the file drops. That single check separates the local tools from the ones running server-side.

Frequently asked questions