Merge, split, reorder, and rotate pages; compress a file that's too big to email; convert to and from images; add page numbers, a watermark, or a signature; redact, fill a form, add or remove a password; or turn a scan into searchable text.
One thing worth saying plainly, once: every one of these tools does its actual work (reading the PDF, redrawing pages, running the text reader) on your device, inside your browser. Nothing gets sent anywhere. Load the page, turn off your Wi-Fi, and everything below still works.
Most tools here (page numbers, watermark, sign, crop, merge, split) only ever add a thin layer on top of your PDF or copy pages between documents. The text underneath stays real text: selectable, searchable, unchanged. Three tools work differently, and it's worth knowing which ones before you rely on them.
If a PDF needs to stay text-searchable, run Compress or the image-based Redact last, not first.
One shared page grid backs Merge, Split, Rotate, and Delete Pages. Load one or several PDFs, drag any page into any order, rotate or remove individual pages, then export one merged PDF or a ZIP of split-out files.
PDF to JPG or PNG at a resolution you pick, and the reverse: JPG, PNG, or WebP images (WebP re-encoded to JPEG) combined into one PDF with page size, orientation, and margin controls. Compress PDF rounds out this group.
Watermark, Add Page Numbers (four formats, custom start number), Crop, Sign (draw or type a signature you place on the page), Redact, and Fill PDF Forms, which detects a document's existing fillable fields and writes your typed answers back in, with an option to flatten.
Protect PDF adds a password plus printing/copying restrictions. Remove PDF Password decrypts a file you can already open. PDF to Text pulls out .txt or .md and flags pages that need OCR instead. Scanned PDF to Searchable PDF is the OCR tool above.
Every tool is its own workspace with its own hash link (like #merge-pdf), so you can bookmark the one you use most.
Merge and split accept several PDFs at once; most other tools take one. A damaged or password-protected file gets a plain error, with a link to Unlock if that's what you need.
A single output downloads as a file; a tool that produces several files (Split, or PDF to JPG on a multi-page document) bundles them into one ZIP. No watermark on the output, no account, no cap on how many times you can run a tool.
Sites like iLovePDF or SmallPDF work fine, but their business model shapes the product: daily file caps, a watermark on the output, or redaction and OCR gated behind a paid plan, all while your document sat on their server the whole time. With no server in the loop, there's nothing to build those limits around.