VECTOR DATABASE
pgvector actions
12 actions you can wire into a flow, alongside everything else in the Vector Database integration.
Every pgvector action
- Count DocumentsCount the documents in the store, with an optional metadata filter
- Create Search IndexAdd an approximate-nearest-neighbour index so similarity search stays fast as the table grows
- Create Vector TableCreate a table that stores documents and their embeddings, with the index that makes search fast
- Delete DocumentsRemove documents by ID or by a metadata filter
- Get DocumentFetch a single document by its ID
- Hybrid SearchCombine meaning-based and keyword search, so exact terms and product codes are not missed
- Insert DocumentsAdd documents to the vector store, embedding the text automatically
- Inspect Vector TableSee a table's columns, embedding dimensions, indexes and row count
- List DocumentsBrowse the documents in the store, with an optional metadata filter
- Search DocumentsFind the documents most similar in meaning to a query
- Update DocumentChange a document's text or metadata, re-embedding it automatically
- Upsert DocumentsInsert documents, or overwrite them if they already exist
Also in Vector Database
Put pgvector in a flow
These actions sit on the same canvas as every other integration, so a pgvector step can follow a form, a schedule or a message without any glue code.