Our take
Activerecent releases- Activity
- 54
- Community
- 32
- Issues
- 21
- Pull requests
- 35
+0 stars in the last 30 days
- Slower to merge pull requests than 92% of the projects we track
An actively maintained, feature-rich TigerGraph assistant, but its narrow database support, AGPL-3.0 license, and concentrated contribution base merit deliberate adoption planning.
Good fit if
- Your organization already uses TigerGraph 4.2 or later and wants graph-plus-document question answering.
- You can provide and configure an LLM service, and can deploy with Docker Compose or Kubernetes.
- You want agentic retrieval or MCP tools and can review the project’s AGPL-3.0 obligations.
Look elsewhere if
- You need a graph or vector database other than TigerGraph, or expect a different retrieval method to be officially supported.
- You need a fully supported product without a separate Statement of Work, or cannot own customizations and operational issues.
- Your project cannot accommodate AGPL-3.0 licensing.
All health signals
| Last commit | 2026-10-12 (6 days ago) |
|---|---|
| Commits, last 90 days | 6 |
| Releases, last 12 months | 10 (latest v2.0.3, 2026-10-12) |
| Contributors | 5 (top contributor: 79% of commits) |
| Issues closed, last 90 days | 0 |
| Pull requests merged, last 90 days | 17 (typically merged in 7 days) |
| Project age | 17 months |
Checked on 2026-10-19 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
23Forks
24Change since Nov 17, 2026: 0 stars·0 forks
Overview
TigerGraph GraphRAG is an AI assistant built around TigerGraph, the graph and vector database platform. It addresses two related tasks: querying structured graph data in natural language, and building a knowledge graph from documents so answers can draw on document passages and graph relationships.
It provides a chat interface and APIs, with classic and agentic chat engines. Hybrid search is the officially supported retrieval method; TigerGraph is the only supported graph and vector database. The project is intended for teams prepared to deploy and operate the services and supply their own LLM provider.
Key Features
- Translate natural-language questions into calls to curated TigerGraph queries for structured-data answers.
- Ingest documents and build a knowledge graph with extracted entities, relationships, and community summaries.
- Combine vector search and graph traversal to retrieve context for document-based questions.
- Offer classic retrieval and agentic Planned or Reactive chat styles, with retrieval traces and citations.
- Ingest local files, cloud-stored documents, or Amazon Bedrock Data Automation outputs.
- Configure chunking, extraction, retrieval, prompts, and per-graph chat models.
- Connect external MCP servers as tools for the agentic chat engine.
- Deploy with Docker Compose or Kubernetes and use a built-in chat UI or APIs.
Use Cases
- A TigerGraph operator can let analysts ask questions about structured graph data without manually composing database queries.
- A team with internal manuals or technical documents can build a knowledge graph and provide answers grounded in retrieved source material.
- Developers can add graph-backed natural-language Q&A to an application using the provided APIs.
- Administrators can combine document ingestion and graph queries in one assistant for operational or analytical workflows.
What you need
Detected in the repository
- Python (main language on GitHub)
- A Compose file to run it with its related services
- A test suite and automated checks on GitHub Actions
License in plain words
AGPL-3.0network copyleft
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the license notice
- Share your changes: yes, the full source, also when people only use it over a network (e.g. as a hosted service)
- Includes an explicit patent grant from the contributors:
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
For a quick Docker-based setup, set LLM_API_KEY and run:
curl -k https://raw.githubusercontent.com/tigergraph/graphrag/refs/heads/main/docs/tutorials/setup_graphrag.sh | bash
This quick-start uses TigerGraph Community Edition and the default configuration. See the README for setup with an existing TigerGraph instance, Kubernetes deployment, and configuration details.
Alternatives
- graphrag: Microsoft GraphRAG builds graph-based context from unstructured text, while TigerGraph GraphRAG centers retrieval on TigerGraph and its graph and vector database.
- KAG: KAG combines knowledge graphs and text retrieval for domain-specific, multi-step QA, while TigerGraph GraphRAG is built around the TigerGraph platform.
- utopia: Utopia organizes evolving knowledge in an ontology-guided, bitemporal graph, while TigerGraph GraphRAG focuses on answering questions using TigerGraph.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| graphrag | Python | AGPL-3.0 | 23 | Active |
| graphrag | Python | MIT | 36.4k | Active |
| KAG | Python | Apache-2.0 | 9.1k | Quiet |
| utopia | Rust | Apache-2.0 | 8k | New |
Considerations
- TigerGraph is the only supported graph and vector database, and the project requires TigerGraph DB 4.2 or later.
- You must provide an LLM service and API credentials. The README describes support for multiple providers, but configuration and service costs remain your responsibility.
- Hybrid Search is the officially supported retrieval method. Other retrieval methods and the agentic chat engine are provided as-is for self-service use; official support requires a Statement of Work with the Solutions team.
- The project is licensed under AGPL-3.0, so review its terms for your intended use and distribution.
- Recent commits and releases indicate active development, but measured contributions are concentrated in one contributor. The available health data also reports no issues closed in the recent 90-day period, so review project support and issue responsiveness against your needs.
Found this useful?
Share it with someone who would like graphrag.