Retrieval

Vector search

Vector search retrieves semantically similar items by comparing embedding vectors rather than relying only on exact keyword overlap.

Expanded definition

Vector search maps text, images, audio, or other data into embedding vectors and ranks nearby vectors with similarity metrics such as cosine similarity or dot product. Production systems often combine vector retrieval with keyword search, metadata filters, reranking, and access control. It is a core retrieval method for RAG and semantic search.

Related terms

Explore adjacent ideas in the knowledge graph.

embedding semantic search vector database RAG

Comparisons, tools, and models that connect to this idea.

Azure Openai Vs Amazon Bedrock (comparisons)
Generative Model (glossary)
Claude 3 5 Sonnet (models)
Adversarial Training (glossary)
Generative Adversarial Network Gan (glossary)
Graph Machine Learning (glossary)