postgres

Documentation

Overview

Postgres search: index the sample files into PostgreSQL with the pgvector extension, then run a semantic search. This is the server-side setup; the only difference from the SQLite examples is the two store constructors.

Needs an OpenAI-compatible embedding server on http://127.0.0.1:1234 (e.g. LM Studio) serving EmbeddingGemma, and a running PostgreSQL with pgvector. The pgx driver is pure Go, so no cgo here.

Start the bundled database (Postgres + pgvector) with Docker, then run from the repository root:

docker compose -f test/docker/docker-compose.yml up -d
go run ./examples/postgres