Traditional keyword search relies on literal matching, so missing synonyms is a fundamental weakness. Vector search first maps text into high-dimensional vectors with an embedding model; semantically similar texts end up close together in vector space — searching for what to do about a headache can hit a note that says relieving migraines.
It's the foundation under RAG and long-term memory systems: memory entries and knowledge-base chunks are vector-indexed and recalled by meaning, not by literal wording.
Try it yourself in Enclave
Open it in your browser — no credit card, no install.