Production Ready Framework For LLM Agents
LlamaIndex is a simple, flexible framework for building agentic generative AI applications that allow large language models to work with your data in any format.
Pip install llama-index
npm install llamaindexValue prop
Production Readiness
The most production-ready LLM framework. State-of-the-art RAG algorithms. Reliable, robust integrations across data loading, indexing, and retrieval. Well maintained integrations and templates.
Flexibility & Customization
The only framework that has infinite layers of customizability for beginner to expert AI engineers. Use our hundreds of integrations, or write your own custom logic with idiomatic, extensible interfaces.
Advanced RAG
Build with cutting edge retrieval techniques, give your LLMs the best context to avoid hallucination.
Agents
Go beyond simple reasoning with agent architecture that can break down complex questions, plan out tasks, and call APIs.
Keep Up With the Latest
Keep up with the latest advancements in search and LLMs and give your LLM app the best performance
Enterprises building with LlamaIndex
Learn to build with LlamaIndex from amazing resources
Get Started with Building Blocks
Build what you want with the components you need
use-cases
From Simple to Advanced
Summarization
Harness the power of AI to distill complex texts into concise, easily digestible summaries. Our advanced algorithms can analyze large volumes of information, ranging from lengthy reports to sprawling data sets, and provide clear, succinct summaries. This tool is perfect for professionals who need to quickly grasp the essence of detailed documents or for anyone seeking to streamline their information intake.
Q&A/Search
The most advanced framework for any Q&A use case and data, from financial analysis to customer support to internal Slack messages. Explore dozens of exclusive techniques for advanced RAG use cases that aren’t available in any other framework.
Agents
Build LLM-powered agents that can perform complex workflows over your data and services. Program them to perform a wide range of tasks, from performing multi-document comparisons to automating your calendar to synthesizing code. Jumpstart your agent with our agent implementations + 30+ tool connectors in LlamaHub or easily write your own.
Loading
Load in 160+ data sources and data formats, from unstructured, semi-structured, to structured data (API's, PDF's, documents, SQL, etc.)
Indexing
Store and index your data for different use cases. Integrate with 40+ vector store, document store, graph store, and SQL db providers.
Querying
Orchestrate production LLM workflows over your data, from prompt chains to advanced RAG to agents.
Evaluating
Evaluate the performance of your LLM application with a comprehensive suite of modules. Measure retrieval and LLM response quality. Effortlessly integrate with observability partners.
TypeScriptBe part of the future of LlamaIndex
Build your career at the forefront of intelligent data systems. Your work shapes how data access will work in the AI era.