Open-source PDF tooling covers several distinct jobs: parsing documents into data, rendering PDFs in a browser, adding OCR to scans, converting files to PDF, editing pages, and signing documents. The right choice depends on which of these you need to automate and where it has to run. Below is a roundup of the options in the Stackora catalog, grouped by what they do.

Libraries for parsing and rendering PDFs

If you are building software that reads PDFs, start with the libraries.

Docling is an MIT-licensed Python library that parses PDF, DOCX, PPTX, HTML and many other document formats into structured data. It is aimed at generative AI pipelines, so its output is meant to feed retrieval and LLM workflows rather than to be displayed to people. The project has about 68,900 stars on GitHub and is published by docling-project.

PyMuPDF is a high-performance Python library for extracting, analyzing, converting and manipulating PDF and other documents, built on the MuPDF engine. Note that it is licensed under AGPL-3.0, which has implications if you distribute or host software that includes it. It is published by pymupdf.

pdf.js is a PDF reader library written in JavaScript. It parses and renders PDF files in the browser using HTML5 and is licensed under Apache-2.0. Use it when the PDF must be displayed inside a web page. It is published by mozilla.

OCR for scanned documents

Scanned PDFs contain images rather than text, so they cannot be searched or copied until OCR is applied.

OCRmyPDF is a command-line tool that adds a searchable OCR text layer to scanned PDFs and produces PDF/A documents. It is licensed under MPL-2.0 and is a good fit for batch jobs and scripts, since it runs as a CLI rather than a service. It is published by ocrmypdf.

Conversion and PDF processing services

If your application has to turn other formats into PDF, a conversion service can remove the need to manage rendering engines yourself.

Gotenberg is a Docker-based API for converting documents to PDF, including HTML, URLs, Markdown and Office files. It is MIT-licensed and saves you from installing and maintaining Chromium or LibreOffice on your own servers. It is published by gotenberg.

Stirling PDF is an open-source PDF platform for editing, merging, converting, signing and redacting PDFs. It runs in the browser, on desktop, or on your own server, so it covers both end-user and self-hosted use. The project has about 94,300 stars and is published by Stirling-Tools. The catalog does not list a license for it, so check the repository before adopting it.

Signing

Documenso is an open-source document signing application and a self-hostable alternative to DocuSign. It is licensed under AGPL-3.0 and is suited to teams that want e-signature workflows under their own control. It is published by documenso.

Document management and archives

Paperless-ngx scans, indexes and archives paper documents into a searchable online archive. It is licensed under GPL-3.0. Docspell addresses a similar need, organizing piles of documents from scanners, e-mails and other sources, and is licensed under AGPL-3.0. Both are self-hosted web apps, so they fit IT teams that want an internal archive.

Desktop viewers and page editors

For people working with PDFs by hand rather than through code:

  • Sioyek is a fast, GPL-3.0 desktop viewer aimed at textbooks and research papers, with smart jumps to figures and references.
  • PDF Arranger is a GPL-3.0 GTK app for merging, splitting, rotating, cropping and rearranging pages.

AI agents and document output

Kami is an MIT-licensed set of layout rules and templates that let AI agents export PDF, PNG and editable PowerPoint. It is categorized under AI Agents in the catalog and is published by tw93. It is useful if your agent produces formatted documents as its output.

How to choose

Start from the job, not the brand:

  • Extracting data for AI or search: use Docling for structured parsing, or PyMuPDF if you need low-level control and can accept its AGPL-3.0 terms.
  • Showing PDFs in a web app: use pdf.js.
  • Making scans searchable: use OCRmyPDF, especially in scripts and pipelines.
  • Generating PDFs from HTML, URLs or Office files: use Gotenberg as a service.
  • Editing, merging and redacting in one tool: evaluate Stirling PDF, self-hosted or in the browser.
  • Self-hosted e-signatures: evaluate Documenso.
  • Archiving paper documents: use Paperless-ngx or Docspell.

Check licenses before you build on any of these. Several are AGPL-3.0 or GPL-3.0, which carry obligations when you distribute or host modified code, while MIT and Apache-2.0 options are more permissive. Review the license against how you plan to deploy.

Recommendation

For a developer building a PDF pipeline, a practical starting point is OCRmyPDF for scans, Gotenberg for conversion, and Docling or pdf.js depending on whether the goal is data extraction or display. Add Documenso or Paperless-ngx only if your workflow needs signing or archiving. Browse the developer tools and libraries catalog pages to compare more options.