GraphRAG (Microsoft)

Modular graph-based RAG. Leiden community detection over knowledge graph; community summaries at multiple levels. DRIFT Search hybrid (Oct 2024).

At a glance

Optimised for: retrieval quality / accuracy on long-context QA

Anti-fit: most are research papers; production fitness varies by maintainer

Most cited RAG-as-memory pattern of 2024 — graph extraction + community summaries gives it macro-level semantic recall most flat-RAG misses.

Indexing cost is high (extraction LLM passes per chunk); operational complexity exceeds vanilla RAG significantly.

v1.0 GA; PyPI distributed. Indexing cost reported at $50–200 per 500 pages.

Other retrieval-as-memory hybrids in the catalog, ranked by inbound references.

RETRO T3
DeepMind. Retrieval-Enhanced Transformer with chunked cross-attention over a 2T-token retrieval database. Frozen BERT retriever + differentiable encoder.
Atlas T3
Meta AI. Jointly fine-tunes retriever + LM for few-shot tasks.
RAPTOR T3
Stanford recursive-summarisation tree over the corpus — clusters and summarises documents into a hierarchical index; retrieves at multiple levels. ICLR 2024.
Self-RAG T4
Trains an LLM end-to-end to emit reflection tokens that decide whether to retrieve, assess passage relevance, and critique output quality — all within a single model.
HippoRAG / HippoRAG2 T3
Hippocampus-inspired RAG. LLMs + knowledge graphs + Personalized PageRank to mimic neocortex/hippocampus roles. NeurIPS 2024.
MemoRAG T4
RAG variant with a long-context global memory model that drafts answer clues used to guide passage retrieval — improves recall in noisy multi-hop QA.