paperless-gpt

paperless-gpt is an integration for paperless-ngx that uses AI models and OCR to generate document titles, tags, and custom fields.

Paperless-gpt pairs with paperless-ngx to automate document organization through artificial intelligence. It extracts text and generates titles, tags, correspondents, and custom fields using providers like OpenAI, Ollama, Google Document AI, Azure Document Intelligence, and Docling Server. Users can review suggestions manually in a web interface or run automated processing flows. The software runs via Docker and supports searchable PDF generation and customizable prompts. It is designed for individuals and teams seeking advanced text extraction and automated sorting for their digital archives.

paperless-gpt at a glance

Best for: Founder, Finance
Runs on: Web, Self-hosted, Open source
Pricing: Free
Released: September 2024
Category: AI productivity tool
Website: github.com/icereed/paperless-gpt

Key features

Uses Large Language Models for OCR
Generates automatic titles, tags, and created dates
Creates automatic correspondents and custom fields
Supports Google Document AI and Azure Document Intelligence
Produces searchable and selectable PDFs
Provides a unified web interface for manual review
Includes customizable prompts via web UI

Use it for

Automating document titles and tags in paperless-ngx
Extracting text from messy or low-quality scans
Generating custom fields and correspondents automatically
Creating searchable and selectable PDF files

Alternatives to paperless-gpt

Paperless-AI — Integrates AI document analysis and RAG chat into Paperless-ngx workflows.
Paperless-ngx — Self-hosted archive that OCRs and tags incoming documents
NuMind — Achieves high accuracy in document extraction and outperforms generalist models.
GenOffice — Free, open-source desktop office suite with built-in AI file editing.
Foglio — It offers a distraction-free environment with ready-to-use document templates.
Unstract — It uses dual LLMs to eliminate hallucinations and ensure extraction accuracy.
Chandra — Provides precise document parsing with bounding box data for complex layouts.
Sparrow — Extracts structured data locally from documents without cloud dependencies.
Instafill.ai — Automates complex form workflows with 25-60 second processing times.
Guizang PPT Skill — Generates structured, professional HTML decks directly inside AI agent workflows.
Rme PDF Scanner — It runs entirely on-device with zero ads, tracking, or backend servers.
WZRD — Combines interactive documents and custom voice agents in one workspace.

All paperless-gpt alternatives