Paperless Ai
An automated document analyzer for Paperless-ngx using OpenAI API, Ollama, Deepseek-r1, Azure and all OpenAI API compatible Services to automatically analyze and tag your documents.
Last verified:
What is Paperless Ai?
Paperless-AI is an AI-powered extension for Paperless-ngx that brings automatic document classification, smart tagging, and semantic search using OpenAI-compatible APIs and Ollama. It enables fully automated document workflows, contextual chat, and powerful customization — all via an intuitive web interface. The tool serves as the most advanced AI companion for Paperless-NGX, transforming document management with powerful AI-driven automation and insights.
Key features include automated document scanning that identifies and processes new documents within Paperless-ngx every 30 minutes by default, AI-powered analysis leveraging OpenAI API and Ollama models (Mistral, Llama, Phi 3, Gemma 2), and automatic metadata assignment for titles, tags, document_type, and correspondent details. The tool offers an interactive Document Chat function powered by Retrieval-Augmented Generation (RAG) that lets you ask natural language questions about your documents and get instant AI-powered responses. Additional features include an AI Playground to experiment with different models, Smart History tracking for all document interactions, and Custom Settings to fine-tune AI processing parameters.
Paperless-AI is designed for Paperless-ngx users who want to automate their document management workflow, individuals and organizations with large document archives who need smart tagging and classification, and anyone who wants to query their document archive using natural language instead of manual filtering. It supports both Automode for fully automated processing and Manual Mode for AI-assisted analysis accessible via the /manual endpoint.
The tool supports a wide range of AI services including Ollama, OpenAI, DeepSeek.ai, OpenRouter.ai, Perplexity.ai, Together.ai, VLLM, LiteLLM, Fastchat, and Gemini (Google), making it highly flexible for different privacy and budget requirements. Users can create predefined processing rules based on existing tags, use selective tag assignment, and assign custom tags to AI-processed documents for easy identification.
Paperless Ai pricing
Pricing model: Freemium
Paperless-AI is free and open-source software licensed under the MIT License. The project is hosted on GitHub and available via Docker Hub as clusterzx/paperless-ai. There are no paid plans or subscription tiers. Users only pay for the AI API costs if they use paid services like OpenAI API, while Ollama and local models are free. The developer accepts support through ko-fi.com/clusterzx, patreon.com/Clusterzx, buymeacoffee.com/clusterzx, and PayPal at paypal.me/bech0r.
Paperless Ai pros
- Automated document scanning every 30 minutes by default
- AI-powered automatic classification and tagging
- Supports OpenAI API and Ollama models
- Compatible with DeepSeek.ai, OpenRouter, Perplexity, Together.ai
- Document Chat with RAG-powered semantic search
- Manual Mode for AI-assisted document analysis
- Modern and intuitive web interface
- Automode for fully automated workflows
- Automatic metadata assignment (title, tags, correspondent, document_type)
- AI Playground to experiment with different models
- Smart History tracking for all document interactions
- Custom Settings for fine-tuning AI parameters
- Predefined processing rules based on existing tags
- Selective tag assignment option
- Custom tagging for AI-processed documents
- Docker support with automatic container restart
- Health monitoring and volume persistence
- Graceful shutdown handling
- Resource management features
- Supports multiple AI models (Mistral, Llama, Phi 3, Gemma 2)
Paperless Ai cons
- Requires Paperless-ngx to be installed first
- Requires Docker and Docker Compose for easy installation
- Must restart container after initial setup for RAG index
- API keys and preferences must be configured manually
- Scan frequency requires manual adjustment from default
- Limited to Paperless-ngx ecosystem only
- Manual scan may show no documents in some bug scenarios
- RAG indexing requires proper container restart
Frequently asked questions about Paperless Ai
How often does Paperless-AI scan for new documents?
Every 30 minutes by default. You can change the scan frequency according to your needs in the configuration settings.
Which AI models are supported?
OpenAI API, Ollama, and all OpenAI-compatible services including DeepSeek.ai, OpenRouter.ai, Perplexity.ai, Together.ai, VLLM, LiteLLM, Fastchat, and Gemini (Google). Many more compatible services may work as well.
What is the best model to use with Paperless-AI?
It depends on your preferences. If you trust OpenAI and their privacy policy for the API, gpt-4o-mini is recommended. If you prefer local processing, llama3.2 or phi4 are the recommended choices.
How do I install Paperless-AI with Docker?
Run: docker pull clusterzx/paperless-ai:latest, then docker run -d --name paperless-ai --network bridge -v paperless-ai_data:/app/data -p 3000:3000 --restart unless-stopped clusterzx/paperless-ai. Access it at localhost:3000.
Do I need to restart the container after setup?
Yes, if you're installing Paperless-AI for the first time, you must restart the container after completing the setup routine where you enter API keys and preferences. This ensures all services initialize correctly and the RAG index builds properly. This step is not required when updating an existing installation.
What is the Chrome extension for?
The Chrome extension is available as a ZIP file at https://github.com/clusterzx/paperless-ai/blob/main/paperless-ai-chrome-extension.zip. Download it and add it to Chrome via the extension manager.
What is Manual Mode?
Manual Mode allows you to manually analyze documents with AI support through a modern web interface. It is accessible via the /manual endpoint and gives you control over which documents to process.
What is the RAG-powered Chat feature?
RAG (Retrieval-Augmented Generation) enables semantic search across your full document archive and provides human-like answers instantly. You can ask questions like 'When did I receive my electricity contract?' or 'How much did I pay for the last car repair?' even without remembering exact titles, senders, or dates.
What are the requirements for Paperless-AI?
You need Paperless-ngx installed and Docker & Docker Compose for the recommended installation method. The tool connects to Paperless-ngx via API using a token from your Paperless-NGX user profile.
How do I create predefined processing rules?
You can specify which documents to process based on existing tags through the Custom Settings. This is optional and allows you to control the automation scope by selecting specific tags for processing.