Skip to content
OriginalSoftware

Stirling PDF: 50+ PDF tools that never leave your server

Stirling PDF puts 50+ PDF tools on your own server — merge, split, OCR, redact, sign — so your documents never get uploaded to a stranger's website.

A flat Swiss-modernist illustration: a strict grid of blank white document sheets, one folded into a small red house shape with a long flat shadow, and a small yellow square accent
One sheet in the grid folds itself into a house — the document that stays home.
On this page

In brief

  • Stirling PDF is a self-hosted PDF platform with 50+ tools — merge, split, OCR, redact, sign, convert — running on your own machine via Docker, a desktop app, or your own server.
  • Latest release v3.3.0 (Nov 5, 2026) adds Stirling Office Convert, a new conversion engine in beta, with better redaction, lower memory on large files, and Windows ARM64 support.
  • The core is MIT-licensed, but parts — the engine and proprietary editor code — are proprietary, and the free plan caps at 5 users and 1,000 processing units a month.
  • OAuth2/OIDC single sign-on is free; SAML2 SSO, audit logging, clustering, and external storage sit on the paid Team and Enterprise plans.
  • Treat the Docker quick start like any self-hosted service: keep it on your network or behind authentication, not open to the internet.

Every PDF tool you have ever used has asked for the same thing: upload the file. Merge a report, sign a contract, strip metadata from a scan — first the document leaves your machine, travels to somebody else's server, and comes back changed. Stirling PDF is the open-source answer to that bargain. It puts more than 50 PDF tools in a browser tab running on your own machine, and the file never has to leave it.

That is the whole pitch, and it is a good one. Stirling PDF is a self-hosted PDF platform: a web interface with tools to merge, split, reorder, rotate, compress, convert, OCR, redact, sign, watermark, and extract pages, plus a REST API covering nearly all of them. You run it as a desktop app, in the browser, or as a server on your own hardware. The current release, v3.3.0, shipped on November 5, 2026 — nearly 97,000 stars on GitHub, more than 6,500 commits, 194 releases.

What it actually does

The tool list is the reason to care. The obvious ones are all there: merge PDFs, split by page range, extract pages, rotate, reorder, add page numbers, compress. Then the ones that usually cost a subscription: OCR a scan into searchable text, redact text and images, add a digital signature or a certificate, flatten forms, strip metadata, convert between PDF and other formats — Office documents, HTML, Markdown, ebooks to PDF, and PDF out to HTML, Markdown, XML, EPUB, and AZW3.

The v3.3.0 release shows where the project is heading. Its headline change is Stirling Office Convert, a new conversion engine still in beta, now available as the converter for most format conversions; the plan is to switch it on by default and then drop the LibreOffice dependency. The same release brought better redaction, merge and split improvements, lower memory use on large files, a JPDFium upgrade, Windows ARM64 support, and faster mobile scanning.

There is also an automation side most PDF websites never offer: no-code pipelines for running the same sequence of operations over many files, and a REST API exposing nearly every tool, so a script can do in one call what would otherwise be an afternoon of clicking. The documentation frames this directly at agents: "Let Claude, GPT and other agents run Stirling tools for you."

Running it

The quickest way in is one Docker command:

text
docker run -p 8080:8080 docker.stirlingpdf.com/stirlingtools/stirling-pdf

Then open localhost:8080. The documentation also offers a desktop editor for Windows, Mac, and Linux that works offline, and a hosted Stirling Cloud option that needs nothing installed — which, note, puts your files back on somebody else's server, the exact thing the self-hosted route avoids. The interface is available in more than 40 languages.

It is a Java application behind a web UI, so it is heavier than a small native utility — the release notes' mention of lower memory on big files hints that large documents have been a pressure point. On a home server it is fine; on a small VPS, check the memory before you commit.

The quick start puts a file-processing service on a network port. Treat it like any self-hosted service: keep it on your local network or behind a reverse proxy with authentication, and do not leave it open to the internet. Username/password login and OAuth2/OIDC single sign-on are both on the free plan — but the default Docker run is a starting point, not a finished deployment.

The open-core caveat

Stirling PDF calls itself open-core, and the repository's license file makes the split concrete. The bulk of the code is MIT-licensed, but several directories — app/proprietary/, app/saas/, engine/, and parts of the frontend editor — each carry their own proprietary license. The core PDF tools are open source; some capabilities sit behind paid plans.

The plans page draws the line more precisely. A self-hosted server runs on the Free plan until you add Team or Enterprise, and Free includes all PDF tools, OAuth2/OIDC single sign-on, and server certificate signing — with limits of 5 users and 1,000 monthly processing units. Team adds the scaling pieces: external database and clustering, S3 or database file storage, Prometheus metrics, and custom PDF metadata. Enterprise holds SAML2 single sign-on, audit logging and usage analytics, database backup and import email notifications, and offline activation.

That is not a reason to avoid it, but read the license before building a workflow around a specific feature. The 50-plus tools cover the everyday work — merge, split, OCR, redact, sign, convert — and the free plan covers them all. The paid line is mostly about scale, operations, and enterprise authentication and audit. For one person on a home server, the free side is the whole product in practice, inside its user and processing limits. For a team deployment, check which side of the line your features fall on first.

How it compares

The honest comparison is not against one rival but against two habits. The first is the subscription desktop app — Adobe Acrobat and its peers — which does all of this and more, for a monthly fee, with your files on your machine. Stirling PDF does not match Acrobat's depth; it does not need to. For the operations most people actually perform on PDFs, the tool list overlaps heavily, and the price is zero.

The second habit is the free PDF website. Those are genuinely convenient, and they are the reason Stirling PDF exists: every upload is a copy of your document on a server you do not control, under a privacy policy you did not read, retained for a period you did not choose. For a boarding pass, that is a fair trade. For a contract, a medical form, or a scan of an ID, it is not. Stirling PDF's argument is that a browser tab's convenience and local processing's privacy should not be a choice you have to make.

Lighter open-source alternatives exist for narrower jobs — PDFsam for splitting and merging, LibreOffice Draw for the occasional edit — but none offers the same breadth in one place with an API on top. Stirling PDF's breadth is the feature: one tool instead of five.

Who should install it

Install it if you handle PDFs regularly and care where the files go: contracts, scans, forms, anything with personal data. It is at its best on a home server or desktop, where the one-command install becomes a permanent utility. It is also worth a look if you automate document work — the API and pipelines turn a tool you use into one you build on.

Skip it, or at least pause, if you only touch a PDF twice a year — the hosted websites are faster for that — or if you need the deep editing Acrobat offers, like reflowing text across pages. And if you run it, run it deliberately: behind authentication, on a network you control. The point of self-hosting a PDF tool is that your files stay yours. The server is the easy part. Keeping it that way is the work.

More from grep.tips

All articles