
Framework for connecting LLMs to your data with advanced RAG
LlamaIndex is an open-source data framework for building production-ready LLM applications with a focus on retrieval-augmented generation (RAG). It provides data connectors for 160+ sources, advanced indexing strategies, and query engines that handle complex multi-step reasoning over your private data.
Import data from 160+ sources including databases, APIs, PDFs, and web pages
Vector indexes, tree indexes, keyword indexes, and knowledge graphs for optimal retrieval
Simple retrieval, multi-step reasoning, and sub-question decomposition for complex queries
Build autonomous retrieval agents that decide how to search and synthesize information
Advanced document parsing for PDFs, tables, and complex layouts
Managed RAG pipeline service for production deployment without infrastructure management
Measure retrieval quality, answer relevance, and faithfulness with built-in metrics
Build semantic search over company documents, wikis, and knowledge bases with RAG pipelines
Create chatbots that answer questions grounded in specific documents with citation support
Build agents that query structured and unstructured data sources to answer complex analytical questions
Combine information from multiple data sources into coherent, grounded answers
Best LangChain alternative for RAG applications — purpose-built data retrieval with faster indexing, better document parsing, and a simpler API for data-centric AI apps
Best for teams where prompt quality depends on data retrieval — RAG applications, document Q&A, and knowledge bases where getting the right context into the prompt matters more than prompt wording.
Works with OpenAI, Anthropic, Google, and other LLM providers through unified interface

The open-source AI-native vector database for search and retrieval