Vector databases
Retrieval substrate for RAG (Pinecone, Milvus, pgvector)
Vector databases (Pinecone, Weaviate, Milvus/Zilliz, Qdrant, Chroma, and pgvector on Postgres) store embeddings and power semantic retrieval for RAG and agent memory. They are the memory tier of the AI application stack, increasingly commoditized into general databases even as specialized players persist.
Players
Pinecone, Milvus, Qdrant, pgvector
Role
embedding retrieval / RAG
How it fits the stack
Vector databases with what it depends on (above) and what it feeds (below). The figure renders as a crawlable diagram and upgrades to an interactive 3D graph as it scrolls into view.
Vector databases in the AI stack. Vector databases with its immediate upstream dependencies (top) and downstream dependents (bottom) in the AI value chain. Hover a node in 3D, or read the full relationships below.
Graph data (text) — 7 entities, 6 relationships
- Agent frameworks (LangChain, LlamaIndex, etc.) —uses→ Vector databases
- Agent Orchestration Frameworks —depends on→ Vector databases
- Agentic AI / autonomous workflows —used by→ Vector databases
- Claude (Opus / Sonnet) —used by→ Vector databases
- LangChain / LangGraph —uses→ Vector databases
- Legal AI —depends on→ Vector databases
Feeds ↓ · 6