For AI agents: The documentation index is at https://docs.digitalocean.com/llms.txt. Markdown versions of pages use the same URL with index.html.md in place of the HTML page (for example, append index.html.md to the directory path instead of opening the HTML document).

DigitalOcean Vector Databases support OpenSearch, PostgreSQL, and Weaviate vector workloads. How you create and configure a cluster depends on the engine:

• Weaviate (public preview): Use the Control Panel, the Vector Databases API at /v2/vector-databases, or doctl vector-databases. See the Weaviate quickstart for the full walkthrough.
• OpenSearch and PostgreSQL with pgvector: Use the Control Panel, the Managed Databases API, or doctl.

You can create vector database clusters from the DigitalOcean Control Panel, the Managed Databases API, or doctl.

Create a Cluster

To create a cluster, choose the engine that matches your workload:

• OpenSearch: Use for search-first vector workloads, hybrid keyword and vector search, remote embedding models through ML Commons, and OpenSearch-based observability or full-text search workloads. To create one, see Create an OpenSearch Vector Database Cluster.
• PostgreSQL: Use when you want to store embeddings alongside relational data and query vectors with SQL. PostgreSQL vector databases use pgvector and support Standard and Advanced editions. To create one, see Create a PostgreSQL Vector Database Cluster.
• Weaviate: Use for vector-native workloads that need a purpose-built vector database. To create one, see Managed Weaviate.

To compare engines, see Choosing Between OpenSearch, Weaviate, and PostgreSQL with pgvector.