LlamaIndex starts from the data. Loading, chunking, embedding, indexing and retrieval are the first class ideas, and agents are built on top of that. If the hard part of your problem is getting the right context in front of the model, this is built for it.
Reach for it when
- Your agent answers from a corpus of documents
- Retrieval quality decides whether it works at all
- You want readers and indexes rather than writing ingestion yourself
Look elsewhere when
- Your agent mostly calls APIs and does not read documents
- You already have a retrieval stack you are happy with
What the lessons will cover
- 01Install and setup
- 02Loading documents
- 03Chunking and nodes
- 04Embeddings and vector stores
- 05Query engines
- 06Retrievers and rerankers
- 07Agents over an index
- 08A document agent
Others in data and rag
HaystackPipelines for search and retrieval heavy applications.
DSPyStop writing prompts. Compile them instead.Common questions
Is LlamaIndex free to use?
LlamaIndex is open source and free to run yourself. You still pay whichever model provider you point it at, and this tutorial is free with no signup.
Do I need to know Python to use LlamaIndex?
Basic Python is enough. Functions, dictionaries and imports cover most of what LlamaIndex asks of you.
When should I not use LlamaIndex?
Your agent mostly calls APIs and does not read documents. You already have a retrieval stack you are happy with.
Lessons for LlamaIndex are being written. Meanwhile the LangGraph tutorial covers the same ground: state, tools, loops, memory and human approval. Most of it carries straight over.
Start the LangGraph tutorial →