Paperless-AI is an AI-powered extension for Paperless-ngx that automatically analyzes and tags documents using OpenAI-compatible APIs and Ollama.
The tool solves the problem of manual document organization in large archives by automating classification, tagging, and retrieval. It detects new documents in Paperless-ngx, analyzes their content using configurable AI backends, and automatically assigns titles, tags, document types, and correspondents. Beyond automation, it provides a retrieval-augmented generation interface for semantic search and natural language queries across the full document archive, allowing users to ask contextual questions like "When did I sign my rental agreement?" rather than relying on keyword matching.
Paperless-AI suits users who maintain large document collections in Paperless-ngx and want to reduce manual tagging overhead. It works with multiple AI providers including Ollama for local processing, OpenAI, DeepSeek, Gemini, and other OpenAI-compatible services, giving adopters flexibility in choosing between cloud and self-hosted models. The tool includes a web interface for manual review of sensitive documents and supports custom rules to control which documents are processed and what tags are applied. The README notes that official AI integration is coming to Paperless-ngx itself, which may affect the project's future relevance.
The project shows substantial real-world adoption, with most open issues raised by external users rather than the core team. Maintainers typically respond to new issues and pull requests within hours. However, the README contains an important notice that the repository is not currently maintained; the maintainer is rewriting the codebase with a more stable architecture but is limited to evenings and uncertain whether the rewrite will be completed given the upcoming official Paperless-ngx AI integration. Work in the issue tracker centers on bug reports, completed items, and investigation of reported problems.
