Paperless-ngx — Self-Hosted Document Management with OCR

What it is

Paperless-ngx is an open-source document management system that digitizes, OCRs, indexes, and archives both physical and digital documents. Drop a scanned document or PDF into the system, and it automatically extracts text via OCR, applies tags, assigns correspondents, and makes everything searchable. It runs self-hosted via Docker.

This tool is for individuals and organizations who want to go paperless without relying on cloud document services. It is also useful for developers building document processing pipelines.

How it saves time or tokens

Paperless-ngx automates the tedious process of organizing documents. The OCR engine extracts text from scanned images, making them searchable without manual data entry. Machine learning-based auto-tagging learns your categorization patterns and applies them to new documents. For AI workflows, the full-text search API provides document retrieval that can feed into RAG systems.

How to use

1. Deploy Paperless-ngx via Docker Compose.
2. Configure consumption directories for document ingestion.
3. Drop documents into the consumption folder.
4. Search and manage documents through the web UI.

# Clone and start with Docker Compose
git clone https://github.com/paperless-ngx/paperless-ngx.git
cd paperless-ngx/docker/compose

# Copy environment template
cp docker-compose.env.example docker-compose.env

# Start the stack
docker compose up -d

# Create admin user
docker compose exec webserver python3 manage.py createsuperuser

# Access at http://localhost:8000

Example

API usage for document search:

import requests

api_url = 'http://localhost:8000/api'
headers = {'Authorization': 'Token your-api-token'}

# Search documents
response = requests.get(
    f'{api_url}/documents/',
    headers=headers,
    params={'query': 'invoice 2026'}
)

for doc in response.json()['results']:
    print(f"{doc['title']} - {doc['created_date']}")
    print(f"Tags: {doc['tags']}")

Related on TokRepo

• Self-hosted solutions — More self-hosted tools
• Document tools — Document processing and management

Common pitfalls

• OCR quality depends on scan quality. Ensure documents are scanned at 300+ DPI for reliable text extraction.
• The initial setup requires Docker and basic Docker Compose knowledge. The configuration file has many options.
• Storage grows with your document collection. Plan disk space for both originals and generated thumbnails.
• Auto-tagging requires training data. The ML classifier needs at least 10 documents per tag before it starts making useful suggestions.
• Paperless-ngx does not handle encrypted PDFs. Decrypt them before ingestion.
• Review the official documentation before deploying to production to ensure compatibility with your specific environment and requirements.
• Start with default settings and customize incrementally. Changing too many configuration options at once makes debugging harder.
• Keep your installation updated to the latest stable version. Security patches and bug fixes are released regularly.