Back to Arena

Neo4j LLM Graph Builder

by Neo4j Labs

System Card

OrganizationNeo4j Labs
Released2024-02
Architecturegraph-rag / Unstructured-to-Neo4j ETL + chat
DetailsFastAPI + React app that extracts entities/relations from PDFs, videos, web pages into Neo4j 5.23+. Schema-based extraction, multi-LLM (OpenAI/Gemini/Anthropic/Ollama), chat QA with source attribution.
Parameters—
Domainrag-retrievalknowledge-graph
Open SourceYes
WebsiteVisit
CodeRepository
neo4j-labsetlcypherbloom-visualization

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
2/5
RULER
72.359p
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
0/2
LoCoMo
no data
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
73.333p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
74.270p
HotpotQA
76.882p
Agent Task Memory
0/1
AgentBench-Mem
no data
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
1/1
RAGAS
73.385p
Sources:Neo4j LLM Graph Builder (neo4j-labs/llm-graph-builder); evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)Neo4j LLM Graph Builder (neo4j-labs/llm-graph-builder); evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)Neo4j LLM Graph Builder (neo4j-labs/llm-graph-builder); evaluated on RAGAS: Automated Evaluation of Retrieval-Augmented Generation (Exploding Gradients, 2309)Neo4j LLM Graph Builder (neo4j-labs/llm-graph-builder); evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)Neo4j LLM Graph Builder (neo4j-labs/llm-graph-builder); evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)Neo4j LLM Graph Builder (neo4j-labs/llm-graph-builder); evaluated on RULER: What's the Real Context Size of Your Long-Context Language Models (NVIDIA, 2404)