Chunking
Chunking is splitting documents into smaller passages before embedding them, so retrieval returns a relevant section rather than an entire file.
Chunk boundaries determine what can be retrieved. Splitting mid-sentence or mid-table destroys meaning; splitting on headings preserves it. Overlapping chunks slightly reduces the chance that the answer falls across a boundary.
Related terms
- Embedding
- An embedding is a list of numbers representing a piece of text’s meaning, so that texts with similar meanings sit close together in that numeric space.
- RAG (retrieval-augmented generation)
- RAG is a technique that retrieves relevant documents from an external store and places them in the prompt, so the model answers from that material rather than from memory.
- Vector database
- A vector database is a store designed to hold embeddings and find the ones most similar to a query embedding quickly.
More on retrieval and agents
See the full glossary, read the guides, or put it into practice in the prompt builder.