TAI BUI
← Glossary
Glossary

What is Chunking?

Breaking text into segments before embedding for retrieval. Chunk size determines the granularity of search results. Too small: loses context. Too large: dilutes relevance. Common strategies: fixed-size with overlap, sentence-based, or semantic splitting. Typical chunk size: 256-512 tokens with 10-20% overlap.

What people say

Splitting documents into pieces

Why it's called that