Quickstart
Run EigenSearch locally with Docker Compose and index your first documents.
Run a full EigenSearch stack (app, workers, and Postgres) with Docker Compose.
Requirements
- Docker and Docker Compose
- An LLM API key (or a self-hosted model endpoint)
Run
EigenSearch is distributed as a Docker image and runs with Docker Compose. From your deployment directory:
cp .env.example .env # set LLM_API_KEY
docker compose up -dThe app comes up on http://localhost:8080. Postgres (with pgvector) starts in the
same stack. Nothing else to provision.
Index a document
curl -X POST http://localhost:8080/api/v1/ingest \
-H "Authorization: Bearer $EIGENSEARCH_TOKEN" \
-F "file=@handbook.pdf"Documents are chunked, embedded, and indexed for lexical + vector search. Ask a question from the UI or the REST API; answers come back with citations.
Next
- Connect a real source → Connectors
- Enforce source permissions → Security