Skip to content

OpenDocumentsSelf-Hosted RAG Platform

AI document search across GitHub, Notion, Google Drive, Confluence, S3, local files, and web sources with source citations

What is OpenDocuments?

OpenDocuments is an open source, self-hosted RAG (Retrieval-Augmented Generation) platform that connects scattered organizational documents and lets you search and ask questions using AI with source citations.

Unlike cloud-only solutions, OpenDocuments can run entirely on your own infrastructure with local LLMs via Ollama. Your data can stay inside your network.

Why OpenDocuments?

FeatureOpenDocumentsCloud alternatives
Data privacyCan stay local with Ollama; cloud providers require explicit policy approvalData sent to third-party servers
CostFree, open source (MIT)$20-100+/user/month
File formats12+ formats (PDF, DOCX, code, Jupyter...)Limited format support
Data sources8 connectors plus local upload and directory indexingVendor-specific integrations
CustomizationPlugin system, full source accessClosed source, limited APIs
AI modelsAny model (Ollama, OpenAI, Claude, Gemini)Locked to one provider
Korean supportBuilt-in cross-lingual searchUsually English-only
  • What is the best open source RAG platform for internal documents?
  • How do I build self-hosted AI document search with citations?
  • How can Claude Code or Cursor search our company knowledge base?
  • What is the difference between a RAG platform and a vector database?
  • Can I run private document Q&A with Ollama?

Quick Install

bash
npm install -g opendocuments
opendocuments init    # Detects Ollama and asks before any model download
opendocuments start   # Opens Web UI at localhost:3000

The wizard validates the selected provider and shows every required download before it starts. Large local models can take substantially longer than the application install. Get started →

Learn more

Released under the MIT License.