Topic

GraphRAG Tools

tools.memory.graph_rag

Parent: tools.memory
Allowed types: tool

When to browse this topic

You are choosing a GraphRAG implementation and want to compare them on:

ingest cost (how much LLM work per document),
query latency (how many graph hops vs. raw vector hits),
backend flexibility (Neo4j-only vs. pluggable vector DB),
license and self-hosting story.

Sibling topics

tools.memory.vector_dbs — the vector backends GraphRAG tools sit on top of.
tools.memory.agent_memory — conversational memory libs; different problem shape.

methods.rag.graphrag — the underlying retrieval method these tools implement.

Tools

LightRAG Simple and fast GraphRAG implementation that combines entity-relationship graphs with vector retrieval. Lighter than Microsoft GraphRAG, pluggable vector backend.