PDF tools · Private by design
Merge, split & compress PDFs without uploading them
Combine PDFs into one file, cut a document into page-range files, or squeeze a bloated scan down to size — right here in your browser. Your files never leave your device, so contracts, statements and scans stay exactly where they belong.
Your PDFs, processed here
Pick a job, add files, download the result. Nothing is uploaded — see how it works and the questions below.
Drop PDF files here, or press Enter to choose them
Files are read locally in this tab. They are never uploaded.
Files
No files yet. In Merge mode add at least two; Split and Compress use the first file in the list.
How to use it
- Pick a job. Merge combines several PDFs into one. Split cuts one PDF into page-range files. Compress makes a smaller copy, losslessly or by re-rendering pages as images.
- Add your files. Drag PDFs onto the page or use the picker. They are read locally, listed with size and page count, and can be reordered or removed.
- Set the options. A name for the merged file, ranges like 1-3, 5, 7-9 for splitting, or a resolution and quality for image compression.
- Download the result. Processing happens in the tab with a progress bar. Before-and-after sizes are shown for compression, and every output is a fresh file.
How it works — and what it cannot do
Local by architecture, not by promise
When you add a file, the browser hands it to this tab as bytes in memory. pdf-lib (MIT) reads, copies and writes the PDF structure for merging, splitting and lossless compression, and Mozilla’s PDF.js (Apache-2.0) renders pages to images for scan compression. Both libraries are served from this site and cached by the service worker like other first-party assets, so after the first load the tool works with no connection at all. There is no upload endpoint to trust: the network tab stays empty, and airplane mode is the easiest audit.
Which compress mode should you choose?
- Text documents, invoices, exports: lossless rebuild. Expect small savings at best — these files are usually already compact — but text, links in content streams and vector sharpness survive.
- Scans and photo-heavy PDFs: image compression. 150 DPI at ~70% JPEG quality is a good email-friendly balance; drop to 110 DPI when size matters more than fine print.
- Anything you must keep searchable: lossless only, or re-run OCR on the image-compressed result elsewhere.
Honest limits
- Encrypted (password-to-open) PDFs are rejected with an explanation, not silently mangled.
- Merging copies pages, not document outlines: bookmarks, form fields, layers and attachments may be lost. Always keep originals until you have checked the output.
- Image compression is one-way. It cannot make text searchable again, and on simple text pages it can produce a larger file — the result panel says so and lets you keep the original instead.
- Everything runs on your device, so thousand-page files take real time and memory. The progress bar reports page by page.
Why “no upload” matters for PDFs in particular
PDFs are where sensitive paperwork lives: contracts, bank statements, IDs, medical bills, signed forms. A server-side tool necessarily holds a copy of that document, however short its retention promise. Processing locally removes the whole category of risk — no copy to leak, subpoena or retain — and as a bonus there are no queues, size caps sold as upgrades, or watermarks. If a job genuinely needs server-class OCR or repair, use a desktop tool you control rather than a web upload.
Questions
Are my PDF files uploaded to a server?
No. Merging, splitting and compressing all run locally in this browser tab using JavaScript (pdf-lib and PDF.js, loaded from this site). Your files are read into tab memory only, are never sent over the network, and disappear when you close the tab. You can verify this by loading the page, going offline, and processing a file — it still works.
How do I merge PDF files into one?
Choose Merge, add two or more PDFs (drag and drop or the file picker), put them in order with the up and down buttons, name the result and choose Merge PDFs. Pages keep their original size, text and quality; the output is a new file, so your originals are untouched.
How do I split a PDF by page ranges?
Choose Split, add one PDF, and write ranges such as 1-3, 5, 7-9. Each range downloads as its own PDF named after the original plus its pages. Choose “every page separately” to get one single-page PDF per page. Ranges are 1-based and are checked against the file’s real page count before anything downloads.
What is the difference between the two compress modes?
Lossless rebuild copies every page into a fresh, tightly packed PDF and strips document metadata. Text stays selectable and quality is unchanged, but savings are modest unless the original was bloated. Image compression re-renders each page as a JPEG at your chosen resolution and quality, which can shrink scans and photo-heavy PDFs by 50–90%, but the text layer and vector sharpness are lost. If the “compressed” file comes out larger, the tool says so plainly.
Why did my PDF not get smaller?
Many PDFs — especially ones exported by modern office software — are already well compressed, so a lossless rebuild can save almost nothing, and re-rendering simple text pages as images can even make them bigger. Compression pays most on scanned documents and PDFs with large embedded photos. Try the image mode at a lower resolution, and keep whichever result is genuinely smaller.
Is there a file size limit?
There is no server limit because there is no server. The practical limit is your device’s memory: files are held in the tab, and very large PDFs (hundreds of megabytes, or thousands of pages in image-compression mode) can be slow or run a phone out of memory. Everyday documents of a few dozen megabytes are fine on typical hardware.
Can it open password-protected or damaged PDFs?
Encrypted PDFs that need a password to open are not supported — remove the password in a desktop PDF reader first, with the owner’s permission. Mildly damaged files sometimes open after the parser’s repair attempts; if a file cannot be read, the tool tells you which file failed instead of downloading something broken.
Does it add a watermark or change my pages?
No watermark is added, ever. Merge and split copy pages exactly as they are. Compress in image mode intentionally re-renders pages (that is how it shrinks scans), and the tool warns you before it does. Your original files on disk are never modified — every result is a new download.
Tools that pair with this one
Same desk, different job. All free, all in the browser.