Panna
Processed locally · nothing uploaded

How it works & privacy

Your files never leave your device

When you edit, compress or merge a PDF here, the file is read directly from your device and processed inside your browser using WebAssembly and JavaScript. It is never uploaded to a server. You can confirm this yourself: open your browser's developer tools, watch the Network tab, and you'll see no document being sent anywhere. The browser loads app files, fonts and OCR assets from this site. Anonymous counters and optional feedback make separate requests, described below.

Saved drafts stay in your browser

There are no accounts, and Panna does not keep server copies of your documents. The PDF editor and Organize workspace save drafts in your browser's local storage database, called IndexedDB. A draft includes the original PDF, edits, text still being entered, view settings and custom fonts you add. Drafts can remain after you close or reload the page so you can resume editing on this device. Workspace drafts include source PDFs, applied document versions, page order, transformations, recent undo history, generated outputs, tool settings and uploaded tool assets. Pending images can be saved before creating a PDF. A draft can retain the original content even after you redact or clean a generated PDF; discard the draft when you no longer need that recovery copy. Passwords are never saved. No document bytes are included in reusable recipes; recipes store only your chosen operations and layout.

Use “Clear saved draft” in the editor to remove that stored document and pause autosave, or “Discard” on the editor or workspace recovery screen to remove a saved draft. Clearing this site's browser data also removes drafts. Private sessions, storage limits and browser cleanup can prevent saving or erase drafts, so download documents you need to keep.

Other tools offer an explicit “Save session copy” action. It stores input files, supported settings and available outputs in IndexedDB. Resume or delete these copies from “Saved sessions”. Continuing between tools temporarily stores the selected PDF on this device; the receiving tool removes the transfer after opening it. Passwords are never included in saved settings. Your device also keeps your light-or-dark-mode preference and the date we last counted your visit. Those preferences contain nothing from your files.

Anonymous usage statistics

To learn which tools are worth improving, Panna keeps simple tallies in its own database: how many times each page was viewed, roughly how many people visited each day, and how many times each tool produced a download. These are plain counters — the digital equivalent of a tally mark. They include no files, no file names, no cookies, and no identifiers of any kind, so there is no way to tell which visits were yours. We use no third-party analytics service; the counts go straight to Panna's own database. In your Network tab this appears as one small request to firestore.googleapis.com carrying nothing but counter increments — never your document. The database is hosted on Google Firebase, so as with any web request Google's servers technically see your IP address while delivering it — but Panna stores no IP addresses and no identifiers of any kind. And if your browser's Do Not Track setting is on, the counter is skipped entirely.

Feedback you choose to send

The feedback form works the same way: the text you type is sent — only when you press send — to Panna's own database, with nothing attached but the date. No name, no email, no identifiers, so we can't tell who sent it or reply. Because it's stored as written, please don't include personal details in your message.

OCR and fonts load from this site

The OCR and Make editable tools use Tesseract, an open-source text-recognition engine. Its worker code, WebAssembly core, and English, German, French, Spanish, Hindi, Arabic and Hebrew language data are served from this site itself — not a third-party CDN — so no outside service ever receives a request tied to your document. The first run downloads those assets from us (they're then cached by your browser), and the recognition itself happens entirely on your device, like everything else here.

The editor's bundled Noto fonts are also served from this site. TTF and OTF fonts you choose from your device are read locally and included in local editor drafts; they are not uploaded. Your browser may cache downloaded fonts and OCR assets for later use. Password protection uses a locally hosted QPDF WebAssembly module; password verification and encryption also run on your device.

What each tool does

Panna currently offers 42 tools. Every one of them works the same way: your file is read on your device, processed in your browser, and never uploaded. Tools marked “beta” are newer and still being refined.

Organize

  • PDF workspace (beta) — Arrange pages, combine tools, undo changes and resume later.
  • Merge PDFs — Combine documents and arrange their pages visually.
  • Split PDF — Mark split points or create named page groups.
  • Extract pages — Choose pages visually and save just what you need.
  • Rotate pages — Turn selected pages and preview the result.
  • Resize & crop — Drag crop edges or fit pages to an exact paper size.
  • Pages per sheet — Arrange 2 or 4 pages on A4 or Letter sheets.
  • Remove pages — Select unwanted pages, remove them and undo if needed.
  • Continuous page — Create scrolling strips with adjustable widths and gaps.
  • Review & clean up — Review blank, duplicate and sideways page suggestions.
  • Split by chapters — Turn bookmarks into named PDFs, including opening pages.

Convert

  • Images → PDF — Arrange, rotate and fit images onto custom-sized PDF pages.
  • PDF → Images — Choose pages, image quality, transparency and contact sheets.
  • Extract text — Preview readable text, choose reading order and copy or download.
  • Extract images — Browse embedded images, filter duplicates and choose what to save.
  • OCR (make searchable) (beta) — Recognize seven languages, review uncertain words and keep original pages.
  • Recognize scans & edit (beta) — Review which pages need OCR, then continue into the editor.
  • PDF → RTF — Export Unicode text with basic headings, font sizes and emphasis.
  • PDF → Markdown — Recover headings, lists and paragraphs with a text preview.
  • PDF → XML — Export text, page coordinates and a versioned XML schema.

Edit

  • Compress PDF — Choose a size target and compare quality before downloading.
  • Watermark — Preview text watermarks, repeated tiles and selected-page templates.
  • Page numbers — Preview page, Roman or Bates numbers with range and margin controls.
  • Header & footer — Place names, dates and page tokens with reusable settings.
  • Sign PDF (beta) — Place drawn, typed or uploaded signatures, initials and dates.
  • Fill PDF forms (beta) — Fill and clear detected fields; keep them editable or flatten on purpose.
  • Edit PDF (beta) — Edit text, review replacements, draw, undo and resume local drafts.
  • Redact (beta) — Review redaction areas and verify removal from rebuilt pages.
  • Edit metadata — Inspect existing metadata, then keep, replace or clear chosen fields.
  • Flatten PDF — Choose fields and annotations to bake into visible page content.
  • Remove annotations — Remove selected annotation types while keeping form fields.
  • Overlay PDFs — Choose base and overlay, map pages and preview alignment.
  • Add image / stamp — Drag, resize and repeat image stamps with page-accurate previews.
  • Unlock forms — Choose read-only fields to make editable; inspect the result.
  • Scanned appearance — Preview photocopy effects, paper tone and reproducible grain.
  • Adjust colours — Compare brightness, contrast and grayscale on selected pages.
  • Repair PDF — Try structural repair, inspect the report or choose a visual rebuild.
  • Reusable workflows — Save a sequence of watermarks, numbering, headers and compression.

Security

  • Clean hidden data — Review and remove hidden data; verify the saved output.
  • Protect / unlock PDF (beta) — Encrypt with AES-256 or remove encryption using the current password.
  • Inspect PDF — Explore pages, metadata, forms, attachments and document structure.
  • Inspect scripts & actions — Review embedded scripts and action locations without running them.

Compress's optional “smallest” mode rebuilds pages as images, which removes selectable text, and Merge doesn't retain document bookmarks from multiple source files. Workspace exports explain when interactive form fields or annotation appearances must be flattened, when accessibility tags are lost, and when an unsupported appearance prevents export. Visual text replacement and whiteout in the editor leave the underlying content in the PDF and do not securely remove sensitive information.

Limits

Because everything runs in your browser, very large files are limited by your device's memory. Files up to 100 MB each are supported. Use Protect / unlock PDF with the current password to unlock an encrypted document before editing it. Previewing an encrypted output asks for its password without saving that password.

← Back to tools