Tool overview
LightRAG is listed under AI Infrastructure & MLOps AI tools.
What is LightRAG?
LightRAG combines knowledge-graph relationships with vector retrieval, incremental indexing, multimodal parsing, configurable storage, a WebUI, and a REST API server for self-hosted RAG applications.
Best for
Engineering teams building self-hosted graph RAG systems
Who is it for?
Decision note
Rebuilt from the original export under the complete V412/V411 factual-source-verification workflow. Preview only. Apply remains blocked until Failures = 0, Warnings = 0, Unmapped = 0, Missing = 0; explicit clears are reviewed; image import is disabled; a representative WordPress edit screen is compared with the export and proposed row; and a post-Apply zero-change Preview succeeds.
Key features
Graph and vector hybrid retrieval
Incremental insert, update, and delete workflows
Multimodal parsing with MinerU and Docling
WebUI, REST API, Docker, and multiple storage backends
Use cases
Build graph-enhanced RAG
Index changing enterprise knowledge
Process multimodal documents
Explore entities and relationships visually
Pros
- Free and MIT licensed
- Self-hosted with broad storage choices
- Active development and large community
Limitations
Version 1.5.0rc3 is a release candidate
1.4.16 is used as the stable snapshot. Output quality depends on extraction, chunking, embeddings, and graph configuration.
Pricing details
Billing options
Pricing note
LightRAG is free under MIT. Users pay for the models, storage, and infrastructure they connect.
Supported languages
- English
- Chinese
- Korean
- German
- Ukrainian
- Russian
- Japanese
- Vietnamese
Integrations
Ollama
OpenAI-compatible models
OpenSearch
Neo4j
PostgreSQL
MongoDB
Langfuse
RAGAS
MinerU
Docling
Please log in to join the discussion.