Merge PDFs
Merge PDFs copies pages from each file into one output in the order you set — vector stays vector when it can; no upload.
Merge, split, compress, convert, edit, and secure PDFs in this tab with pdf-lib and PDF.js. Nothing is uploaded.
Runs in your browser · Nothing uploaded
PDF tools are the other half of “please don't upload that.” Contracts, scans, offer letters, and filled forms are exactly the files people hand to random websites for a merge or a compress. This category keeps the bytes in the tab: libraries load from this origin, work runs in memory, you download a result or you close the tab.
Start with the everyday jobs — merge, split, rotate, organize, compress — then convert when you need images or a best-effort trip through Word, Excel, PowerPoint, Markdown, CSV, or EPUB. Security tools include password protect and a redaction path that rasterizes pages so text does not survive under a black box. Edit PDF text is the live click-to-type editor; OCR is for scans that never had a text layer.
Honesty beats a support ticket: PDF→Word/Excel/PowerPoint reconstruct text, they do not clone layout. Compress makes pages into images on purpose. Sign stamps an image, it is not certificate e-sign. Compare is side-by-side eyes, not an automatic redline. If a tool cannot do a thing well offline, it is not advertised as if it could.
Merge PDFs copies pages from each file into one output in the order you set — vector stays vector when it can; no upload.
Split PDF extracts a contiguous page range or zips one single-page PDF per page for offline filing.
Compress PDF re-renders pages and JPEG-encodes them at Light/Medium/Heavy — smaller files, usually no selectable text afterward.
Organize pages shows PDF.js thumbnails so you can reorder or drop pages, then save a new PDF locally.
Rotate PDF turns every page by 90°, 180°, or 270° without re-rasterizing the content.
Crop & resize trims uniform margins in inches or sets page boxes to A4/Letter/Legal — resize does not reflow text.
PDF to JPG/PNG renders each page in the tab; one page downloads as an image, many pages as a zip.
Images to PDF embeds JPEG/PNG with pdf-lib — one image per page by default, or stack vertically on one page.
Word to PDF is text-first via mammoth into a simple Helvetica layout; complex Word art will not match.
PDF to Word builds a best-effort .docx from the text layer — expect cleanup, not a perfect round-trip.
Excel to PDF prints the first sheet as a landscape table; long cells truncate for layout.
PDF to Excel heuristically groups positioned text into rows and columns — made for real tables, not novels.
CSV to PDF parses with Papa Parse and prints the same style of offline table PDF.
Markdown to PDF runs marked, then lays plain text into a simple PDF — not a themed docs printer.
PowerPoint to PDF extracts .pptx slide text onto widescreen pages; images and animations are skipped on purpose.
PDF to PowerPoint builds one text slide per page from the text layer — restyle afterward.
PDF to EPUB packs the text layer into a reflowable EPUB 3 book (one chapter per page), not a fixed facsimile.
Add watermark overlays centered text or a PNG/JPEG on every page with opacity and rotation.
Add page numbers stamps a start index at bottom left, center, or right on every page.
Headers & footers stamps repeating plain-text lines at fixed top and bottom positions.
Extract text copies the PDF text layer to the page or a .txt — scans need OCR first.
Edit PDF text: click a paragraph and type; auto-OCR for scans; whiteout + redraw on download — cover-up, not surgical delete.
Encrypt PDF password-protects via pdf-lib with restrictive permissions; we do not invent a fake AES marketing number.
Remove password unlocks a PDF when you already know the password — not a crack tool.
Redact PDF paints black regions and rebuilds image-only pages so underlying text cannot be selected.
Flatten PDF bakes AcroForm fields into static page content for archival sends.
PDF metadata shows author/title/dates/producer and can strip them locally — page content stays put.
OCR searchable PDF runs Tesseract per page and adds an invisible text layer (best-effort alignment).
Sign PDF places a signature image on a page — visual stamp, not certificate e-sign.
Repair PDF best-effort copies salvageable pages from a damaged file; encrypted files need unlock first.
Compare PDFs shows synced side-by-side rendered pages — eyes only, no automatic redline export.
PDF to audio extracts the text layer and speaks it with local speechSynthesis — no MP3 download.
No. Libraries load in the tab from this origin. File bytes stay in memory on your device until you download a result or close the tab.
A cloud LLM version would send document contents off-device. That breaks the site's promise. An on-device model may come later; a Gemini-backed fake will not.
No. PDF stores positioned drawing commands, not office layouts. These converters reconstruct text for editing. For a visual twin, export PDF from the original app or use PDF to JPG/PNG.
Whiteout covers while you edit; text objects can still exist underneath. Redact rasterizes pages after black boxes so there is no selectable text layer left. Use Redact for irreversible removal.
There is no server to upload to. Each page is a static file, and the work happens in your browser. Your file is read from disk into memory, processed, and handed back as a download. Once the page has loaded you can disconnect from the network and it keeps working. The longer explanation