Vector Graph RAG

Graph RAG with pure vector search — no graph database needed, single-pass LLM reranking, optimized for knowledge-intensive domains.

Why Vector Graph RAG?

Most Graph RAG systems require a dedicated graph database (Neo4j, etc.) and complex multi-step retrieval with iterative LLM calls. Vector Graph RAG takes a fundamentally different approach:

flowchart LR
    subgraph Traditional["Traditional Graph RAG"]
        direction LR
        A1[Documents] --> B1[Graph DB\nNeo4j]
        B1 --> C1[Cypher\nQueries]
        C1 --> D1[Iterative\nLLM Calls]
        D1 --> E1[Answer]
    end

    subgraph VectorGraphRAG["Vector Graph RAG"]
        direction LR
        A2[Documents] --> B2[Milvus\nVectors]
        B2 --> C2[Vector\nSearch]
        C2 --> D2[Single LLM\nRerank]
        D2 --> E2[Answer]
    end

Key Advantages

• No graph database — The entire knowledge graph lives in Milvus as vectors. No extra infrastructure, no schema management, no graph query language.
• Single-pass reranking — Unlike agentic approaches (IRCoT, multi-step reflection), we call the LLM just once to rerank candidate relations. Simpler, faster, and cheaper.
• Knowledge-intensive friendly — Designed for domains where dense factual knowledge matters: legal documents, financial reports, medical literature, novels, and more.

Features

No Graph Database

Pure vector search with Milvus — no Neo4j, no ArangoDB, no extra infrastructure to deploy or manage.

Single-Pass Reranking

One LLM reranking call, no iterative agent loops. 2 LLM calls total per query vs 5+ for iterative approaches.

Multi-Hop Reasoning

Subgraph expansion discovers connections across documents for complex multi-hop questions.

Zero Configuration

Milvus Lite by default — works as a local file, no server setup needed. Just pip install and go.

State-of-the-Art

87.8% average Recall@5 on standard multi-hop QA benchmarks, competitive with HippoRAG 2.

Visual Explorer

Interactive frontend with step-by-step retrieval visualization showing how the algorithm reasons.

How It Works

flowchart LR
    Q[Question] --> EE["Entity\nExtraction"]
    EE --> VS["Vector Search\n(Milvus)"]
    VS --> SE["Subgraph\nExpansion"]
    SE --> LR["LLM Rerank\n(single pass)"]
    LR --> AG["Answer\nGeneration"]

    style VS fill:#e3f2fd,stroke:#1565c0
    style SE fill:#e3f2fd,stroke:#1565c0
    style LR fill:#fff3e0,stroke:#e65100
    style AG fill:#e8f5e9,stroke:#2e7d32

1 Extract entities from the question 2 Vector search finds similar entities and relations in Milvus 3 Subgraph expansion traverses the graph to discover multi-hop connections 4 LLM reranking selects the most relevant relations (single pass) 5 Generate answer from the selected context

Quick Example

from vector_graph_rag import VectorGraphRAG

rag = VectorGraphRAG()  # reads OPENAI_API_KEY from environment

rag.rebuild_texts([
    "Albert Einstein developed the theory of relativity.",
    "The theory of relativity revolutionized our understanding of space and time.",
])

result = rag.query("What did Einstein develop?")
print(result.answer)

Ingestion semantics

The legacy add_* ingestion helpers rebuild the full knowledge base and are planned for removal in v1.0.0. Use rebuild_texts(), rebuild_documents(), or rebuild_documents_with_triplets() for full refreshes, and use upsert_documents_by_source() / delete_documents_by_source() when a source file, page, or message changes later. See Incremental Updates for parser integration, source key design, and retry recommendations.

Getting Started

See the Getting Started guide for installation and configuration options.

Performance

Method	MuSiQue	HotpotQA	2WikiMultiHopQA	Average
Naive RAG	55.6%	90.8%	73.7%	73.4%
IRCoT + HippoRAG	57.6%	83.0%	93.9%	78.2%
HippoRAG 2	74.7%	96.3%	90.4%	87.1%
Vector Graph RAG	73.0%	96.3%	94.1%	87.8%

Benchmark Details

Recall@5 on standard multi-hop QA benchmarks. Uses the same pre-extracted triplets as HippoRAG for fair comparison. See Evaluation for full details.