Docspell review 2026: how this free, open-source, self-hosted document organizer with OCR and auto-tagging works, deployment options, and alternatives.
Docspell is a free, open-source, self-hosted document organizer that helps individuals, families, and small groups digitize and organize documents from scanners, email, and other sources with minimal manual filing. It focuses on automatically extracting metadata like correspondent, date, and tags rather than requiring rigid manual folder structures.
The project is maintained by developer Eike Kettner and a community of open-source contributors, distributed under the AGPLv3 license, with no corporate parent, headquarters, or funding behind it.
Documents are OCR'd via Tesseract and analyzed with Stanford CoreNLP-based natural language processing to suggest correspondents, tags, and dates, with the system learning from confirmed metadata over time. All content becomes full-text searchable through the web interface.
Docspell organizes access around shared 'collectives' so multiple users, such as members of a household, can work with the same document pool under their own individual accounts, and everything is accessible via a REST/HTTP API, a mobile-friendly web app, a native Android app, and a command-line interface.
Docspell is entirely free and open source under the AGPLv3 license, with no subscription tiers, feature gating, or user limits imposed by the project itself.
There is no official managed cloud hosting offered by the Docspell project; users self-host via Docker Compose, native packages, or building from source, with the only costs being their own server and storage infrastructure.
Yes. Docspell is completely free and open source under the AGPLv3 license, with no paid tiers, subscriptions, or usage limits.
Docspell is created and primarily maintained by developer Eike Kettner along with a community of open-source contributors, with no corporate backing.
Yes, Docspell is designed to be self-hosted, typically via Docker Compose or a native package, so all documents and data stay on infrastructure the user controls.
Yes. It uses Tesseract OCR to extract text from scanned images and PDFs, making them fully searchable.
Docspell places more emphasis on machine-learning-assisted metadata suggestion and supports a multi-user 'collective' model for households or small groups, whereas Paperless-ngx has a larger community and more mainstream single-archive adoption.
Yes, Docspell offers a native Android app for capturing and uploading documents, alongside its mobile-friendly web interface.