Open Glean

Open Glean is a self-hosted AI workspace for searching files, notes, saved webpages and knowledge synced through Hydra DB connectors. Its browser app runs on your machine or server, including in Docker, but retrieval uses the remote Hydra DB API by default and requires a Hydra key. You can change the Hydra backend endpoint. Local answer models require explicit enablement and are disabled by default.

Questions produce streamed answers with inline citations and a panel of sources. You can limit searches to selected databases and collections, or search within one collection. Search works without an answer model; generated responses require one.

Answer generation accepts OpenAI-compatible endpoints, including Ollama and LM Studio for local inference or cloud providers through OpenRouter. Choosing a local model doesn't remove the default remote retrieval dependency. If you choose a cloud model, the retrieved context used for its answer goes to that provider. Optional web search uses OpenRouter's web plugin and adds web citations beside Hydra sources.

Deep Research breaks longer questions into sub-questions, gathers branch findings and produces an answer with a combined citation list. A timeline displays its progress. The app also manages Hydra connectors and displays Hydra's knowledge graph as a mindmap.

Open Glean is open source under Apache 2.0. Conversations persist in MongoDB when available, with browser localStorage as a fallback. The server uses API credentials held in encrypted session cookies.