#4646th of 51 in RAG, knowledge and search
llm-graph-builder
Turns PDFs, web pages and videos into a Neo4j knowledge graph
- Stars
- 5.3k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Jul 2026
- Language
- Jupyter Notebook
4,589 on 21 Apr 20265,304 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
A FastAPI backend and React frontend that ingest local files, S3 or GCS objects, YouTube, Wikipedia and web pages. An LLM called through LangChain extracts entities and relationships, which are written to Neo4j. It previews graphs in Bloom and offers chat over the graph with several retrieval modes. It is an experimental Neo4j Labs project, not a supported Neo4j product.
Who it is for: Teams building Neo4j knowledge graphs from documents for graph RAG
Strengths
- Ingests local files, S3, GCS, YouTube, Wikipedia and web pages into one graph
- Chat modes include vector, graph, fulltext, entity_vector and global_vector, with source metadata
- Custom node and relationship schema can be set before extraction
- Extraction works with OpenAI, Gemini, Diffbot, Anthropic, Bedrock, Groq and Ollama
Weaknesses
- Experimental Neo4j Labs project, not part of Neo4j's supported products
- Requires Neo4j 5.23+ with APOC; no other graph store
- Docker Compose not supported with Neo4j Desktop; run backend and frontend separately
- Most non-OpenAI providers are labelled dev deployed version
What it needs
- GPU optional
- Compose
- Needs Neo4j 5.23+ with APOC, LLM API key (OpenAI, Diffbot, etc.) or Ollama, YouTube transcript proxy key
- Models: OpenAI, Gemini, Diffbot, Anthropic, Azure OpenAI
Quick answers
Is llm-graph-builder open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run llm-graph-builder with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does llm-graph-builder need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in RAG, knowledge and search
See all 51| Rank | Project | Score |
|---|---|---|
| 1 | RAGFlowRAG engine with deep document parsing, template chunking and agentic retrieval | 90 out of 100 |
| 2 | LightRAGGraph-based RAG framework with a REST API server and web UI | 90 out of 100 |
| 3 | deeptutorSelf-hosted AI tutor with knowledge bases, guided learning, books and quizzes | 89 out of 100 |
| #4 | Open NotebookSelf-hosted NotebookLM alternative for research notebooks, chat, and podcasts | 89 out of 100 |
| #5 | MaxKBEnterprise knowledge-base agent platform with RAG, workflows and MCP tools | 87 out of 100 |