Home / Jobs / Vector database

Vector database

The best vector database for RAG and semantic search is pgvector because it scores highest on our published formula (78.3/100): Adds vector search to the Postgres you already have, with no new service.

The job: “I need semantic search or RAG retrieval over embeddings.” · Last checked Oct 19, 2026

Pick

pgvector

Adds vector search to the Postgres you already have, with no new service.

Not for: Very large or specialized vector workloads.

Pricing: Open source Postgres extension. source, checked 2026-10-19

Visit pgvector ↗

78.3

/ 100

Runner-up

Qdrant

Open-source vector database with a free cloud tier.

Not for: Small apps already on Postgres (see pgvector).

Pricing: Open source. Cloud free tier: single node, 0.5 vCPU / 1 GB RAM / 4 GB disk. source

78.1

/ 100

Also considered

Chroma (60.5/100): Easy open-source embedding store for prototypes.

Score breakdown

Tool	Fit (editorial) ×0.4	Adoption (GitHub stars) ×0.25	Momentum (release freshness) ×0.2	Attention (HN, 30d) ×0.15	Score
pgvector	4/5 → 0.75	23.4k★ → 0.819	Oct 1, 2026 → 0.798	211 pts → 0.789	78.3
Qdrant	5/5 → 1	35.1k★ → 0.852	Oct 5, 2026 → 0.841	0 pts → 0	78.1
Chroma	3/5 → 0.5	29.5k★ → 0.837	Oct 17, 2026 → 0.978	0 pts → 0	60.5

Missing signals are dropped and the remaining weights re-normalized. See methodology. Signals fetched Oct 19, 2026.

FAQ

What is the best vector database for RAG and semantic search?

What is the best alternative to pgvector?

Our runner-up is Qdrant (78.1/100). Open-source vector database with a free cloud tier. Not for: Small apps already on Postgres (see pgvector).

Who should not use pgvector?

Very large or specialized vector workloads.

How much does pgvector cost?

Open source Postgres extension. (checked 2026-10-19, source: https://github.com/pgvector/pgvector)

Related jobs

Parse PDFs and docs for RAG: pick Docling
Web scraping for LLMs: pick Firecrawl