Glossary term
Glossary term
Memory and Retrieval
Indexing organizes and stores data so AI can quickly search and retrieve it, ensuring documents, transcripts, or knowledge articles are structured for fast, accurate information retrieval.
Elasticsearch and OpenSearch build inverted indexes optimised for BM25 retrieval.
Pinecone, Weaviate, and Qdrant build vector indexes (HNSW, IVF, ScaNN) for semantic search.
LlamaIndex provides high-level abstractions for indexing documents into vector and tree structures.