# Semantic Retrieval Semantic retrieval finds records by meaning rather than requiring exact keyword overlap. Modern systems often use embeddings and vector indexes, sometimes combined with lexical, metadata, or graph constraints. The language engine used explicit symbolic and distributional representations rather than neural embeddings. Those indexes can now participate in semantic or hybrid retrieval while retaining interpretable reasons for each match. Related: [[wiki/Hybrid Retrieval|Hybrid Retrieval]], [[wiki/Vector Retrieval|Vector Retrieval]], [[wiki/Distributional Semantics|Distributional Semantics]], [[wiki/Retrieval-Augmented Generation|RAG]]. Source: [OpenAI — Vector stores](https://platform.openai.com/docs/api-reference/vector-stores)