AI Document Search
Ask questions in natural language across GitHub, Notion, Google Drive, Confluence, S3, and local files. Get cited answers with source links — not hallucinations.
Self-hosted AI document search — connect your scattered docs and ask questions with source citations
OpenDocuments is an open source RAG (Retrieval-Augmented Generation) tool that connects your scattered organizational documents and lets you search and ask questions using AI — with source citations so you know exactly where the answer came from.
Unlike cloud-only solutions, OpenDocuments runs entirely on your own infrastructure with local LLMs via Ollama. Your data never leaves your network.
| Feature | OpenDocuments | Cloud alternatives |
|---|---|---|
| Data privacy | Runs locally, data never leaves your network | Data sent to third-party servers |
| Cost | Free, open source (MIT) | $20-100+/user/month |
| File formats | 12+ formats (PDF, DOCX, code, Jupyter...) | Limited format support |
| Data sources | 10+ connectors (GitHub, Notion, Drive...) | Vendor-specific integrations |
| Customization | Plugin system, full source access | Closed source, limited APIs |
| AI models | Any model (Ollama, OpenAI, Claude, Gemini) | Locked to one provider |
| Korean support | Built-in cross-lingual search | Usually English-only |
npm install -g opendocuments
opendocuments init # Auto-detects Ollama, pulls models
opendocuments start # Opens Web UI at localhost:3000Three commands. Under 5 minutes. Get started →