Pinecone Vector DB
Managed (SaaS) vector database for dense vector / semantic retrieval at scale, maintained by Pinecone. Also supports native sparse/term search, enabling Hybrid Search from a single instance.
- Website: https://www.pinecone.io/
Notes
- Used as the vector retriever alongside a Lucene engine in a multi-retriever Hybrid Search setup, with Metarank fusing the result sets via Learning to Rank — see Hybrid Search and Learning-to-Rank with Metarank.
- Pinecone’s Learn blog is a frequent source of search/IR explainers in this vault.
Related Tools
- Qdrant Vector DB · Weaviate Vector DB · Milvus Vector DB — alternative vector databases