An alternative to iLovePDF and Smallpdf

On those services the work happens on a server. Your document is transmitted, written to disk somewhere, processed, and held until a retention timer deletes it. Their privacy policies describe this accurately and the retention windows are usually short — the point is not that they are dishonest, it is that the file left your machine and you are now trusting a policy.

Here the work happens in the browser tab you already have open. There is no upload step, so there is no retention question, no queue, and nothing to delete. You can verify it rather than take it on trust: open your browser's network panel and process a file, and nothing carrying the document goes out.

Everything below is a consequence of that single architectural difference, in both directions.

No account, because there is no server holding one. No daily task limit, because nobody is paying for the compute — your device is doing the work. No file-size ceiling beyond what your own machine can hold in memory, and no upload wait, which on a large file is often longer than the processing.

No watermark, ever, on any output at any size. This is the one that most often sends people looking for an alternative, and it is worth being precise: nothing here stamps anything onto your document. The only tool that adds a watermark is the one whose job is adding one, using text you chose.

And it keeps working with the connection off. Once the page has loaded, the tools and their processing engines are stored on your device — useful on a train, and a hard guarantee that nothing is being sent.

Format conversion fidelity is the real one. Converting a complex Word document or a formatted spreadsheet to PDF well means running a full layout engine — effectively a word processor — and those services run exactly that on their servers. Our conversions read the content and lay it out cleanly rather than reproducing the original's exact appearance. For a CV or a report that is fine; for a document whose formatting is the point, they will do better.

The same applies in reverse to PDF-to-Word: reconstructing an editable document from a PDF is guesswork, and more compute usually means better guesses.

They also offer things that require a server by definition — shared links, cloud storage integrations, team accounts, e-signature workflows with an audit trail. If you need a countersigned document with a legal audit trail, you need a service that keeps records, and that means a server.

Finally, on a very old or very low-memory device, a large job that a server would absorb may struggle in a browser tab.

If the document contains anything you would not email to a stranger — a contract, a CV with your address on it, a bank statement, a medical letter, anything with an identity number in it — do it on your own machine.

If you need pixel-exact conversion of a heavily formatted file, or a signature workflow with a legal record attached, use a service built for that and accept the trade.

Most everyday PDF work is merging, splitting, compressing, rotating and converting images, and none of that needs a server at all.

Do not take our word for it either. Open the developer tools in your browser, switch to the Network tab, and run a file through any tool on this site. You will see the page and its code load, and then nothing — no request carrying your document, because the policy the page is served with forbids it.

The same test is worth running on any tool that claims to be private. It takes thirty seconds and it is the only claim in this category that can be verified from outside.