Paperless-ngx

Capabilities

• Scan-to-searchable OCR pipeline Tesseract-based text recognition, plus thumbnail generation, runs over incoming scans so paper documents become full-text searchable records rather than opaque images.[1][2]

• Consumption-folder intake Anything dropped into the monitored consume folder is imported automatically, which makes a network scanner or scanner-side upload the core hands-off workflow.[1][2]

• Email and mobile capture Mail fetching rules pull attachments from designated accounts, and official apps for iOS and Android add documents on the go alongside browser upload.[1][2]

• Tags, correspondents, and types Every document can carry tags, a correspondent, and a document type, and matching rules can assign them automatically, which keeps large archives navigable.[1]

• Tantivy full-text search The search index covers recognized text and metadata with saved views and advanced filtering, so retrieval is a query rather than a folder hunt.[1][2]

• Multi-user permissions User accounts with object-level permissions let a household or small team share one archive while restricting who sees or edits what.[1]

• Documented REST API A versioned, OpenAPI-described API covers documents, tags, saved views, and more with token authentication, which supports scripts, dashboards, and third-party integrations.[1]

• Docker and bare-metal self-hosting The recommended Docker Compose setup ships with defaults, PostgreSQL is the recommended database, and bare-metal installation is documented for operators who prefer it.[1][2]

Who will love it

• Paper-heavy households going digital: if the goal is scanning years of letters, receipts, and manuals into something searchable, the consume-folder plus OCR workflow matches that job directly.[1][2]

• Self-hosters who want data at home: the whole stack runs on your own Linux machine or NAS with local text recognition, so documents never need to leave your network.[1][2][3]

• Small teams needing a shared paper trail: tags, correspondents, saved views, and per-object permissions give a small office a workable shared archive without per-seat fees.[1]

Who should avoid it

• Anyone who cannot self-host should look elsewhere: there is no vendor-run cloud plan, so if nobody will maintain updates and backups, a managed service fits better.[1]

• Windows- or macOS-only server operators are excluded: the server is documented for Linux via containers or bare metal, with no installer for other desktop operating systems.[1]

• Teams needing editing or e-signature workflows should avoid it: Paperless-ngx archives and retrieves documents rather than editing, collaborating on, or signing them.[1]

Fit and compatibility