Glossary term
Glossary term
Memory and Retrieval
Database optimized for embedding similarity search.
Pinecone hosts Notion's RAG embeddings at scale - billions of vectors across millions of user workspaces - serving sub-100ms semantic search queries that power Notion AI's document retrieval.
Weaviate powers Zeta Global's customer-data platform, storing dense vectors for 235 million consumer profiles to enable real-time semantic audience segmentation for advertising campaigns.
pgvector (PostgreSQL extension) is used by Supabase customers to build RAG applications on top of existing PostgreSQL databases, eliminating the need for a separate vector store for projects under 1M rows.