Four PDF jobs, one hub. Fix a sentence in a contract, merge twenty photos into a single document, turn meeting notes into clean pages, or send a CAD drawing to a client who has never opened AutoCAD — all inside your browser tab, where the file already is.
Each tool does one job and hands the result straight back to you. Nothing is queued, quoted or held for review.
Click any line of text to replace it in place, drag a box over anything you need hidden, add new text or drop in a corrected image. The exported file keeps everything else untouched, and you also get a photo-repair bench for flat, noisy scans.
Combine JPG, PNG or WebP files into one document, reorder them by dragging, and set the page size before exporting. Scanned receipts, photo series and screenshots come out as a single tidy file instead of a folder of attachments.
Paste or type your content, choose the page setup, and export a clean multi-page document with real selectable text. Useful for cover letters, handouts, short reports and anything that should look the same on every screen.
A built-in DXF parser writes lines, arcs, circles and polylines as true vector geometry, so the PDF stays razor sharp at any zoom instead of becoming a screenshot. Text entities come across as real, selectable PDF text.
All four are free, unlimited and anonymous. There is no account to create, no "two free exports per day" counter, and no email address required to download a result.
Most PDF work is not one big project — it is four small ones that turn up in the same week. You receive a PDF with the wrong date on it, you photograph ten pages of a signed form, you need to hand over notes that are not a screenshot, and someone asks for the floor plan "as a PDF please".
Keeping those four jobs on one page means one privacy model to learn rather than four sets of terms to read. Whichever tile you click, the file is opened with your browser's own File API, processed by JavaScript in that tab, and handed back to you as a download. There is no upload step in the middle, because there is no server waiting to receive anything.
Three steps, and step two happens on your own processor.
No drag-and-drop onto someone else's infrastructure: the file picker talks to your operating system and nothing else. If the PDF you need to edit is already on this machine, that is where it stays.
Rendering uses pdf.js, the same engine Firefox uses to display PDFs; writing uses pdf-lib. Both are JavaScript running in your tab, so sensitive pages are drawn and rebuilt locally rather than shipped somewhere for processing.
Export produces a new file — your original is never modified. No watermark, no "create an account to download", no waiting for an email link that expires in 24 hours.
The well-known online converters are good products and we are not pretending otherwise. The difference is not the quality of the output — it is where the bytes go in the meantime.
| The tools in this hub | Typical upload-based converter | |
|---|---|---|
| Where the file is processed | Your device | A remote server |
| Upload required | Never | Always |
| Sign-up before download | No | Often, on the paid tier |
| File size cap | Your RAM only | Free tier limits apply |
| Watermark on output | None | Sometimes on free plans |
| Usable with no internet | Yes, once loaded | No |
| Copy retained elsewhere | No copy exists | Usually stored briefly |
| Price | Free | Free tier paid for with file limits |
To be fair, uploading has advantages: a remote server will convert a 500 MB file on a phone that could never do it alone, and cloud tools often do things local JavaScript cannot, such as reliable OCR or heavy compression. If your document is genuinely public — a conference agenda, a published datasheet — any reputable converter will do the job well. The calculus changes for anything you would hesitate to hand to a stranger.
A short field guide to the services people usually have bookmarked, what each one is good at, and the one trade-off to be aware of. We link to them because knowing the difference helps you choose, not because they are bad — every one of them is a mature product.
The best-known all-rounder, with dozens of tools under one interface and a clean, obvious layout — genuinely the benchmark this page borrows its structure from.
Trade-off: files are uploaded to their servers to be processed. Fine for everyday documents, worth thinking twice about for anything confidential.
Strong batch handling and a browser-friendly flow for merge, split and compress jobs; its API is widely used in other people's software.
Trade-off: also server-side, which means waiting on the upload for large files and trusting their retention window.
The reference implementation of the format itself, so fidelity on complex documents is as good as it gets, and conversion of tricky layouts rarely breaks.
Trade-off: conversions run in Adobe's cloud, and the free tier is designed to lead you towards a subscription.
A downloaded application rather than a website, so files are processed locally — the closest in spirit to what these four tools do, plus OCR and form filling.
Trade-off: free usage is limited per day, and it is software you install and update instead of a tab you open.
Excellent at image-side work such as resizing, HEIC conversion and compression before a picture ever becomes a PDF page.
Trade-off: images are uploaded too — which is where our own background remover, running fully offline, is a better fit.
Descriptions reflect each product's own published feature pages and are our honest reading of them; check the source before relying on specifics. We have no affiliation with any of them.
This is not a marketing flourish — there is no backend to this site's tools at all, which makes the privacy promise structural rather than contractual. If nobody receives your document, nobody can leak it, lose it, or be compelled to hand it over.
The honest trade-offs: heavy work uses your own CPU, so a 40 MB drawing is slower on a five-year-old phone than on a laptop. State lives in your browser's local storage and will not follow you to another device unless you export it. And a few tools load an open-source library or a small model from a public CDN the first time you use them — the page says so whenever that happens.
No upload means no copy to delete later, no retention window to read, and nothing to lose in a breach. The strongest privacy policy is the one with no data to describe.
Work begins the instant you pick the file instead of after a transfer. Small jobs feel immediate, and there is no queue behind you.
Once a page has loaded you can drop the connection and keep editing, merging or converting — handy on a plane or behind a locked-down office network.
If you are not sure which tile above fits, this is usually enough to decide.
Wrong date, outdated price, a name misspelled: open it in the editor and click directly on the text to replace it in place.
Drag a box over it with the cover tool. The area is painted over in the exported file — see the redacting guide for how to do it properly.
Merge the photos into one PDF and drag them into the right order first; nobody has to click through ten attachments.
Convert the DXF to true vector output so the client can zoom to 800% and still see clean lines, without installing CAD software.
Type or paste the content and export properly paginated pages with real selectable text rather than a screenshot.
Run it through the repair bench for auto contrast, denoise and sharpening, then place the cleaned page back into the PDF.
No. Each tool reads the file with your browser's File API, renders pages with pdf.js, and writes the new document with pdf-lib — all JavaScript running in your own tab. There is no server component that receives your document, so there is nothing to store or forward.
Yes. No page counters, no "two free conversions today", no export caps. The site is funded by advertising, which is why you may see ads around the tools — never injected into the content you are editing.
Yes, once the page itself has loaded. Reload while you are online, then disconnect: rendering and export continue to work because every library involved has already been cached by your browser.
No artificial one. The real ceiling is your device's memory — a current laptop comfortably handles files well past 100 MB, while an older phone may slow down somewhere above 20 MB.
You can change how it looks, which is usually what matters. A scan has no text layer to select, so use the cover tool to mask the old content and the add-text tool to type over it in any language your fonts support.
None at all — no logo, footer badge or trial notice. There is also no account step between clicking export and having the file.
Nothing remains anywhere. Tools that remember state keep it in your own browser's local storage; clear site data and every trace is gone. Keeping things on your device is a privacy advantage, but it also means there is no cloud copy to recover.
Optical character recognition of scanned pages needs another service, very large-batch processing is impractical in a tab, and password-protected files must be unlocked elsewhere first. Everything else here is a solvable local job.
Longer write-ups on the jobs where doing the work locally actually changes the outcome.
Why dragging a black box can still leave the original words recoverable, and how to cover content so that it is genuinely gone.
What vector output means for a client who will zoom in, and how fidelity compares with a rasterised screenshot.
The privacy and latency difference between an on-device model and an image upload service.
Unit and colour converters, JSON formatting, QR codes, word counts, mortgage comparisons and more — all local.