NLP & Language · intermediate · concept 136 of 311 · Updated Nov 4, 2026
Chunking
Chunking splits documents into pieces small enough to retrieve and fit in a model's context. Chunks that are too small lose meaning, and chunks that are too large bury the answer. Splitting by document structure and adding context to each chunk usually beats fixed sizes, and poor chunking is a common hidden reason a RAG system answers badly.
Key terms
Chunk sizeOverlapStructure-aware splittingContextual retrievalMetadata
Learn these first
Where you meet it in the real world
Document RAG, policy and legal search, technical documentation assistants
Guides and articles