Welcome to LlamaIndex.TS
LlamaIndex.TS is a data framework for building production-ready LLM applications in TypeScript and JavaScript. Use it to integrate large language models with your own data through powerful patterns like RAG (Retrieval-Augmented Generation), agents, and workflows.Quickstart
Get started in minutes with your first RAG application
Core Concepts
Understand the key concepts behind LlamaIndex.TS
Examples
Explore comprehensive examples for all features
API Reference
Browse the complete API documentation
What Makes LlamaIndex.TS Special?
LlamaIndex.TS is designed to be lightweight, flexible, and production-ready for building LLM applications across any JavaScript runtime.Multi-Runtime Support
Run your LLM applications anywhere JavaScript runs:- Node.js >= 20 ✅
- Deno ✅
- Bun ✅
- Nitro ✅
- Vercel Edge Runtime ✅ (with some limitations)
- Cloudflare Workers ✅ (with some limitations)
Browser support is currently limited due to the lack of support for AsyncLocalStorage-like APIs.
Key Features
RAG Made Easy
Build powerful retrieval-augmented generation systems with simple, composable APIs for indexing, querying, and chat.
Agent Workflows
Create sophisticated agentic systems with the modern
@llamaindex/workflow package for multi-step reasoning and tool usage.Modular Architecture
Install only what you need. Provider packages for LLMs, embeddings, and vector stores keep your bundle size minimal.
Multiple LLM Support
Works with OpenAI, Anthropic, Groq, Gemini, Llama, Mistral, and many more providers out of the box.
Supported LLM Providers
LlamaIndex.TS integrates with all major LLM providers:- OpenAI (GPT-4, GPT-3.5)
- Anthropic (Claude)
- Google (Gemini)
- Groq
- Llama2, Llama3, Llama3.1
- MistralAI
- Fireworks
- DeepSeek
- ReplicateAI
- TogetherAI
- HuggingFace
- DeepInfra
Use Cases
LlamaIndex.TS powers a wide range of LLM applications:Question Answering
Build systems that answer questions over your documents, knowledge bases, or databases using RAG.
Conversational Agents
Create chatbots and assistants that can use tools, access your data, and maintain context across conversations.
Document Analysis
Extract insights from PDFs, emails, transcripts, and other unstructured data sources.
Community and Support
Join our growing community of developers building with LlamaIndex.TS:GitHub
Star the repo, report issues, and contribute
Discord
Join our Discord community for support and discussions
Follow @llama_index for updates and announcements
Playground
Try LlamaIndex.TS interactively in your browser
Open Source
LlamaIndex.TS is fully open source and MIT licensed. We welcome contributions from the community!Next Steps
Installation
Set up LlamaIndex.TS for your runtime environment
Quickstart
Build your first RAG application in 5 minutes