Mechanism◆Article
GraphRAG indexing builds a community hierarchy over the entity graph using Leiden clustering and pre-generates a summary report for every community at every level, making the entire collection's content answerable at query time.
GraphRAG's indexing pipeline clusters entities into hierarchical communities and generates a summary report for each, which pre-computes the answers to questions that span the entire corpus. ✦ AI generated
article author · ByteByteGo Newsletter · 2026-08-19 · original ↗
For every community at every level, a language model generates a community report. Each report contains an overview of that community along with its key entities, relationships, and claims... This is the step that makes whole-collection questions answerable. A summary of what a cluster of documents collectively says gets written during indexing, well before anyone asks about it.
Read full article ↗excerpt · fair-use quotation
- ·Leiden clustering builds entity hierarchy
- ·LLM generates report per community per level
- ·Reports capture overview, entities, relationships, claims
- ·Summaries written during indexing phase
- ·Answers corpus-wide questions before queries
- ·Cluster-level insights available instantly
- ·Each report synthesizes cluster content
- ·Spanning questions resolved by report union
- ·No real-time aggregation needed
Around this claim
Context · 3
Standard RAG works by chunking documents, embedding them into vectors, and retrieving chunks whose vectors are closest to the query vector, relying on the assumption that the text answering a question will be similar to the question itself.article author · ByteByteGo Newsletter · conf 80%A knowledge graph stores entities (nouns) and typed relationships between them, each with a description, which preserves the connections between concepts that plain text chunking discards.article author · ByteByteGo Newsletter · conf 80%Standard RAG's similarity-based assumption fails for 'global queries' that require reasoning across a large collection of documents, because the relevant information is distributed, not located in a few text regions that resemble the query.article author · ByteByteGo Newsletter · conf 70%