severian42/GraphRAG-Local-UI

Health v2 · maintenance only

Adoption not part of the score

2321 stars · 294 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A local-first GraphRAG application adapted from Microsoft's GraphRAG, providing a FastAPI backend plus Gradio-based UIs for indexing, prompt tuning, querying, chatting, and 2D/3D knowledge graph visualization. It works with local LLMs and embeddings via Ollama or OpenAI-compatible APIs.

Use cases

• build a knowledge graph rag pipeline with local llms
• visualize a knowledge graph extracted from documents
• chat with my documents using graphrag and ollama
• tune graphrag prompts locally without cloud costs
• run graphrag indexing through a web ui
• query documents with global and local graphrag search

When to choose

• you want GraphRAG without sending data to cloud LLM providers
• you need an interactive UI for indexing, querying, and graph visualization
• you use Ollama or OpenAI-compatible local model servers
• you want to experiment with knowledge-graph-based retrieval

When to avoid

• you need a production-hardened, stable system - the project is in transition with potential instability
• you only need simple vector RAG without knowledge graphs
• you require support beyond a Mac Studio M2 tested environment
• you need enterprise support or frequent maintenance

Facets

application · maturity active

rag llm-inference data-visualization chat-interface web-framework api-framework large-language-models artificial-intelligence data-visualization self-hosted python self-hosted cross-platform graphrag knowledge-graph local-llm ollama gradio fastapi graph-visualization prompt-tuning retrieval-augmented-generation docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="severian42/GraphRAG-Local-UI")