Paperless-ngx
Document management system that transforms your physical documents into a searchable online archive with OCR.
| Image | ghcr.io/paperless-ngx/paperless-ngx:latest |
| Web UI | http://your-server:8000 |
| Config path | {BASE_DIR}/paperlessngx/data |
| Website | docs.paperless-ngx.com |
| Source code | GitHub |
Resource Impact
| Performance | Medium — OCR processing with Redis broker and SQLite database |
| Storage | Medium — Stores original documents and OCR data |
Installation
sh
mithrandir install paperlessngxArchitecture
Paperless-ngx runs as a multi-container stack:
| Container | Description |
|---|---|
paperlessngx_webserver | Main application with web UI (port 8000) |
paperlessngx_broker | Redis broker for task queue |
Data is stored using SQLite (no separate database container needed). Documents are stored in four directories under {BASE_DIR}/paperlessngx/:
data/— Application data and SQLite databasemedia/— Processed documents and thumbnailsexport/— Document export directoryconsume/— Drop files here to be automatically imported
Configuration
| Variable | Description |
|---|---|
PAPERLESS_OCR_LANGUAGE | Language for OCR processing (default: eng). See available languages |
When HTTPS is enabled via Caddy, PAPERLESS_URL is automatically set to the correct HTTPS URL.
Setup
- Open the web UI at
http://your-server:8000 - Create a superuser account: you'll be prompted on first access
- Drop documents into the
consume/directory or upload them via the web UI