Your pgvector Queries Are Doing Sequential Scans — Here's Why

Automated excerpt

On a few hundred rows, a brute-force sequential scan that compares every vector takes milliseconds. You can query the system catalogs to find every vector column that has no index: JOIN pg_class c ON i. indexrelid = c. oid Every row returned is a vector column doing brute-force scans on every similarity search. At minimum, test similarity search against 100,000+ rows.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.