NGPDFs

NGPDFs is a suite of more than fifty PDF tools that run entirely inside your browser. Merge, split, organize, rotate, crop, compress, watermark, fill forms, add page numbers, convert to and from Office formats, extract text, redact, encrypt and run OCR, all without your file ever leaving your device. There's no upload, no account, no email and no watermark on the output, and after the first visit the whole thing works offline as a progressive web app. The tagline on the site is PDF tools that never see your PDFs, and the architecture is built to make that literally true.

The problem it solves is the one hiding behind every free online PDF site. Convenient as they are, most of them work by uploading your document to a server, which is exactly what you don't want to do with a contract, a medical record, a tax form or a client file. NGPDFs takes the opposite position. The site describes itself as privacy-first by architecture rather than by policy, meaning the tools are built so that a server never sees the document, instead of promising to delete it afterwards. The difference shows up in small ways too. There's no account to tie your documents to, no email capture before the download, and no watermark stamped on the output as a nudge to upgrade.

Technically, every standard operation runs in a Web Worker using WebAssembly-ready libraries, with pdf-lib and PDF.js doing the heavy lifting. A file is validated by its magic bytes, handed to the worker as a transferable buffer, processed, verified, and returned to you as a download. The OCR tool runs Tesseract compiled to WebAssembly, with English and Spanish available today because each language model is several megabytes, and it skips pages that already contain text unless you force it. A service worker caches the app shell and processing assets, which is what makes offline use possible. Compression is offered both as a general tool and as presets that aim at specific sizes, which is the form the problem usually takes when an upload form rejects your file.

The catalogue is wide. Page tools cover merging, splitting, organizing, rotating, cropping, flattening and compressing to target sizes. Conversions go from images, Markdown, CSV, Excel, PowerPoint, HTML and ebooks into PDF, and from PDF out to JPG, Word, Excel, PowerPoint, HTML, EPUB and plain text. Security tools include encryption, password removal, manual and automatic PII redaction, metadata removal, a privacy scanner and a fingerprint check. There are also a few unusual ones, such as turning a PDF into audio or into a handwriting-style rendering. Form filling, flattening, headers and footers, page numbers and colour inversion round out the editing side, and a hash tool under diagnostics lets you verify a file's fingerprint.

It's for anyone who handles documents they can't or shouldn't upload, which means lawyers, clinicians, accountants, HR teams, journalists and ordinary people dealing with their own paperwork. It also suits people who work on planes, in the field or on slow connections, since once loaded it doesn't need the network at all. Nothing about it requires technical skill. You pick a tool, drop a file, and download the result. Students and small businesses get the everyday operations, merging a scan into a report or shrinking a brochure for email, without paying for a desktop suite.

What sets it apart is the honesty of the disclosure model. The stated rule is that if a tool can run locally it will, and any tool that genuinely needs the network will say exactly what leaves your device before it does. The only telemetry is a support widget that receives anonymous usage counters like the tool name and a duration bucket, never file names, contents or extracted text. The site keeps a how-it-works page and a diagnostics page for anyone who wants to check the claims. That transparency extends to naming the libraries it's built on, which is rare for a consumer PDF site.

NGPDFs is free, with no paid tier, no sign-up and no watermark. The tradeoffs are the ones that come with doing everything in a browser. Very large files are limited by your device's memory rather than a server's, OCR covers two languages so far, and the operator behind the site isn't named, so support is thin. For everyday PDF work on documents you'd rather keep to yourself, it's hard to argue with the model. The privacy page spells out which categories of tool run locally, from merging and splitting to text extraction, metadata removal and fingerprinting, and states that the only third party in the picture is the support widget.