# LightRAG | ✨Data Intelligence Lab@HKU✨ | Knowledge Base | Instagit

[EMNLP2025] "LightRAG: Simple and Fast Retrieval-Augmented Generation"

GitHub Stars: 30.3k

Repository: https://github.com/HKUDS/LightRAG

---

## Articles

### [How to Configure LightRAG for High Availability with Distributed Storage Backends](/HKUDS/LightRAG/configure-lightrag-high-availability-distributed-storage)

Configure LightRAG for high availability using distributed storage like Redis Cluster, Qdrant, Neo4j, PostgreSQL, or MongoDB with multiple FastAPI replicas for robust performance.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Implement Custom Entity Types in LightRAG Using addon_params](/HKUDS/LightRAG/implement-custom-entity-types-lightrag-addon-params)

Easily implement custom entity types in LightRAG by configuring addon_params. Override default taxonomy without modifying core code for flexible data extraction.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Migrate Existing Vector Databases to LightRAG's Storage System: A Complete Guide](/HKUDS/LightRAG/migrate-vector-databases-lightrag-storage)

Easily migrate your vector databases from Qdrant Milvus or PostgreSQL to LightRAGs storage system. Our guide ensures seamless data validation dimension checks and preserves your valuable data through automatic migration.

- Tags: migration-guide
- Published: 2026-03-23

### [How to Troubleshoot LightRAG Entity Extraction Issues When the Knowledge Graph Is Not Building](/HKUDS/LightRAG/troubleshoot-lightrag-entity-extraction-kg-issues)

Troubleshoot LightRAG entity extraction issues and knowledge graph errors. Learn to fix token-count guards, LLM output, and storage configurations for successful graph building.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Implement Citation Functionality in LightRAG to Trace Sources with File Paths](/HKUDS/LightRAG/implement-citation-functionality-lightrag)

Learn how to implement citation functionality in LightRAG. Trace knowledge sources with file paths by setting include_references=True during queries and passing file_paths during document insertion.

- Tags: how-to-guide
- Published: 2026-03-23

### [JsonKVStorage vs PostgreSQL in LightRAG: Choosing Between File and Database Storage](/HKUDS/LightRAG/lightrag-jsonkvstorage-vs-postgresql)

Discover the differences between JsonKVStorage and PostgreSQL in LightRAG. Choose the right storage for your RAG application development, from local JSON files to robust database solutions.

- Tags: deep-dive
- Published: 2026-03-23

### [How to Use Pipeline Insert (`apipeline_enqueue_documents`) for Incremental Document Indexing in LightRAG](/HKUDS/LightRAG/lightrag-pipeline-insert-incremental-indexing)

Learn to use LightRAG's apipeline_enqueue_documents for efficient incremental document indexing. Asynchronously add new content, deduplicate, and update your knowledge graph without full re-indexing.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Configure chunk_token_size and chunk_overlap_token_size in LightRAG](/HKUDS/LightRAG/configure-chunking-lightrag-text-splitting)

Learn how to configure chunk_token_size and chunk_overlap_token_size in LightRAG for optimal text splitting. Control text granularity and context preservation easily.

- Tags: how-to-guide
- Published: 2026-03-23

### [Best Practices for Large-Scale Dataset Processing with max_parallel_insert in LightRAG](/HKUDS/LightRAG/best-practices-lightrag-large-dataset-processing)

Optimize large-scale dataset processing in LightRAG. Learn best practices for max_parallel_insert, tuning it from 2 to 8 based on your system limits. Maximize performance efficiently.

- Tags: best-practices
- Published: 2026-03-23

### [How to Integrate LightRAG with LlamaIndex for Advanced Indexing](/HKUDS/LightRAG/integrate-lightrag-llamaindex-advanced-indexing)

Integrate LightRAG with LlamaIndex for advanced indexing using LightRAG's adapter. Learn how to leverage LLM and embedding backends for enhanced document retrieval and analysis.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Use the Workspace Parameter in LightRAG for Multi-Tenant RAG Application Isolation](/HKUDS/LightRAG/lightrag-workspace-isolation-multi-tenancy)

Learn how the LightRAG workspace parameter isolates tenant data across KV, vector, and graph stores. Secure multi-tenant RAG applications with this essential feature.

- Tags: how-to-guide
- Published: 2026-03-23

### [How LightRAG's Reranking Integration Works with Jina AI and Cohere APIs](/HKUDS/LightRAG/lightrag-reranking-integration-jina-cohere)

Discover how LightRAG seamlessly integrates with Jina AI and Cohere rerank APIs. Learn about its unified async helpers and token-aware chunking for efficient document processing.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Configure Token Budget Controls in LightRAG: Managing max_entity_tokens, max_relation_tokens, and max_total_tokens](/HKUDS/LightRAG/configure-token-budget-controls-lightrag)

Learn to configure LightRAG token budget controls max_entity_tokens, max_relation_tokens, and max_total_tokens via environment variables, API, or SDK. Prevent prompt overflow.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Implement Document Deletion with Automatic Knowledge Graph Regeneration in LightRAG](/HKUDS/LightRAG/implement-document-deletion-lightrag-kg-regeneration)

Learn to implement document deletion and automatic knowledge graph regeneration in LightRAG. This guide shows how to delete vectors and entities, then rebuild the graph for integrity.

- Tags: how-to-guide
- Published: 2026-03-23

### [How LightRAG's LLM Caching (enable_llm_cache) Reduces API Costs](/HKUDS/LightRAG/lightrag-llm-caching-reduce-costs)

Reduce LLM API costs with LightRAG's enable_llm_cache. This feature stores LLM responses locally, eliminating redundant calls and saving you money on token-based billing.

- Tags: how-to-guide
- Published: 2026-03-23

### [LightRAG Storage Backends: Complete Guide to KV, Vector, Graph, and Doc Status](/HKUDS/LightRAG/lightrag-supported-storage-backends)

Explore LightRAG storage backends for KV, vector, graph, and doc status. Discover 24 production-ready options including JSON, Redis, PostgreSQL, MongoDB, Neo4j, and OpenSearch.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Set Up LightRAG with Ollama for Local LLM Inference and Context Size Configuration](/HKUDS/LightRAG/setup-lightrag-ollama-local-llm)

Learn to set up LightRAG with Ollama for local LLM inference. Configure context size and optimize your local language model performance with this easy guide.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Implement Custom Embedding Functions in LightRAG Using wrap_embedding_func_with_attrs](/HKUDS/LightRAG/implement-custom-embeddings-lightrag)

Learn to implement custom embedding functions in LightRAG using wrap_embedding_func_with_attrs. Seamlessly integrate your models into indexing and retrieval pipelines with this powerful decorator.

- Tags: how-to-guide
- Published: 2026-03-23

### [How LightRAG Performs Knowledge Graph Extraction and Entity Relationship Building](/HKUDS/LightRAG/lightrag-knowledge-graph-extraction-process)

Discover how LightRAG constructs knowledge graphs. Learn about its three-stage pipeline for entity extraction, relationship building, and persistent storage.

- Tags: how-to-guide
- Published: 2026-03-23

### [How to Configure LightRAG with Neo4j for Production Knowledge Graph Storage](/HKUDS/LightRAG/configure-lightrag-neo4j-production)

Learn to configure LightRAG with Neo4j for production knowledge graph storage. Discover automatic connection pooling, isolation, indexing, and full-text search. For HKUDS/LightRAG users.

- Tags: how-to-guide
- Published: 2026-03-23

### [LightRAG Query Modes Explained: local, global, hybrid, mix, naive, and bypass](/HKUDS/LightRAG/lightrag-query-modes-explained)

Explore LightRAG query modes: local, global, hybrid, mix, naive, and bypass. Understand how each mode optimizes entity-centric, relationship-focused, or vector-only retrieval for better search results.

- Tags: deep-dive
- Published: 2026-03-23

### [How LightRAG Dual-Level Retrieval Works: Local vs Global Search Modes Explained](/HKUDS/LightRAG/how-lightrags-dual-level-retrieval-works)

Explore LightRAG's dual-level retrieval. Understand how local and global search modes process queries using distinct vector stores and graph traversal for precise or broad pattern matching.

- Tags: internals
- Published: 2026-03-23