icereed/paperless-gpt
📦 Open Source Projecticereed
Automate document processing in paperless-ngx using LLMs and Vision models for intelligent classification and data extraction.
paperless-gpt acts as a powerful middleware for paperless-ngx, enabling users to harness the power of Large Language Models and Vision-Language Models (VLMs) for document automation. Written in Go, the project is designed for efficiency and seamless integration with existing document management pipelines. It excels at tasks such as automatic document categorization, intelligent tagging, and extracting specific data points from invoices, receipts, or contracts.
The tool supports a wide range of backends, including Ollama for local, privacy-focused processing, as well as various cloud-based LLM APIs. By utilizing Vision models, it bypasses traditional, rigid OCR limitations, allowing the system to 'understand' the context and content of a document rather than just performing character recognition. This results in higher accuracy for complex documents and reduced manual effort. Key features include configurable prompt templates, support for multiple LLM providers, and a modular architecture that allows for easy scaling as document volumes grow.
💡Highlights
- ├─Integrates LLMs with paperless-ngx
- ├─Supports local Ollama & cloud APIs
- └─Automates tagging and classification
🎯For
- ├─Home automation enthusiasts
- ├─Document management administrators
- └─Privacy-focused developers