Vector databases for AI apps
If you are adding RAG, the database choice is usually: stay on Postgres with pgvector, or buy a specialist (Pinecone, Qdrant, Weaviate). Specialists win on ops and filters at scale. Postgres wins when you already run SQL and do not want another bill. This hub is for that fork, not for picking an LLM.
View rankingsBrowse alternativesBrowse by tags
Do I need a dedicated vector DB?
Not for a prototype. pgvector on a database you already have is enough until metadata filters, recall, or ops hurt.
Pinecone vs Qdrant vs Weaviate?
Pinecone is hosted-first. Qdrant and Weaviate you can self-host. Read the current pricing and SLAs on each page. We will not rank them from memory.
Is this the same as a model hub?
No. Models generate embeddings. This category stores and searches them.
All Vector databases for AI apps tools (13)
Open any tool below for details, features, and pricing.
- ElasticsearchDistributed search engine. Not leftover 8.0-only.
- FaissMeta ANN library. A library, not a hosted vector DB or marketing ally.
- MilvusOpen distributed vector DB. Cloud is Zilliz;
- MySQLOracle relational DB. Not a leftover social-media origin story.
- NeonServerless Postgres. Auth/storage/functions are beta extras.
- PostgreSQLOpen relational DB. Not a leftover BSD slogan page.
- PGVectorPostgres vector extension. Not a Faiss wrapper. No hosted dollar seat.
- PineconeManaged vector DB. Not an unpriced ML toolkit.
- QdrantOpen vector search. Not an unpriced modeling tool.
- SupabaseOpen Firebase alternative. Auth/storage shipped; not still-in-development.
- UpstashServerless Redis. Not a Kafka-only unpriced kit.
- ValdCloud-native ANN engine.
- VespaOSS search+vector engine. Cloud has a free trial; unit table exists. Not an unpriced toolkit.