The three steps

  1. 1
    +

    Add your file

    Drop a PDF, image or Office document into the workbench. The browser reads it into memory; no request is made.

  2. 2

    Choose the result

    Pick a tool, enter a known password or set options such as page size, rotation or quality.

  3. 3
    ✓

    Save it locally

    The finished file is offered as a download from memory. Nothing is stored anywhere else.

What runs inside your browser

YaPDF ships the same kind of libraries desktop PDF software uses, compiled for the browser. They download from this domain the first time a tool needs them and are cached by your browser afterwards.

pdf.js

Mozilla's PDF engine, the one built into Firefox, reads pages for PDF to Word, PDF to text and PDF to JPG, and checks passwords before unlocking so a typo produces a clear message.

pdf-lib

Assembles new documents for merging, extracting pages, rotating, watermarking, cleaning metadata and combining files. Pages are copied as PDF objects, so text and vector graphics are never re-rendered.

qpdf in WebAssembly

The command-line tool trusted in professional pipelines, compiled to WebAssembly, removes encryption for Unlock PDF and rewrites files for Fast web view. It runs in a worker thread so the page stays responsive.

Office rendering

Word and PowerPoint files are laid out with docx-preview and pptx-preview, charts are redrawn with ECharts, and each page is rasterised with html2canvas before it is placed in the PDF.

Text shaping and fonts

Text files and watermarks are set with fontkit and self-hosted Noto fonts, so accented Latin, Chinese and Tamil come out as real, selectable glyphs.

What leaves your device

Only requests for the page itself: its HTML, stylesheet, scripts, fonts and the WebAssembly engine, all served from yapdf.yavinesh.com. Your document, its password and the result are never part of any request. You can confirm this in your browser's network inspector while a tool runs.

There are no third-party scripts, fonts, analytics or cookies on this site. The privacy page explains exactly what the web server can see.

Memory, not storage

Files you add live in the memory of the current tab. Results are kept as in-memory blobs until you download them or the tab is closed or refreshed. YaPDF does not use local storage, IndexedDB or a cache of your documents.

Limits worth knowing

  • Scans need OCR. PDF to Word and PDF to text read existing text; image-only pages produce a clear "no selectable text" message.
  • Office files are rendered, not converted. Fonts, complex layouts, equations and embedded objects may differ slightly, and those pages are stored as images.
  • XFA forms are rejected rather than damaged; standard AcroForm fields are preserved.
  • Right-to-left scripts and combined emoji are refused in text files and watermarks instead of being rendered incorrectly.
  • Memory is the only size limit. A hundred-page export works on ordinary laptops; very large files depend on what your browser can hold.
  • Bookmarks (outlines) are not carried over when merging or extracting pages in this release.

Browsers and devices

YaPDF is built on standard web technology and tested most in current Chrome and Edge, including mobile viewports. Recent Firefox and Safari releases are expected to work. If a tool fails in one browser, retry in Chrome or Edge. See the FAQ for troubleshooting.