Qdrant is a self-hostable vector database.

A quick-start guide is here: https://qdrant.tech/documentation/quickstart/

It is useful when combined with n8n as n8n offers a nice integration.

Run Qdrant locally, after pulling it with docker:

docker run -p 6333:6333 -p 6334:6334 \
    -v "$(pwd)/qdrant_storage:/qdrant/storage:z" \
    qdrant/qdrant

Distances

Qdrant offers the following distances:

• Cosine
• Euclid
• Dot
• Manhattan
  Also see here: https://qdrant.tech/documentation/concepts/search/#metrics