Framework for connecting LLMs and agents to your own documents and data.
Retrieval-heavy applications where the hard part is getting your own documents and data into an agent reliably.
Your agent does not read private data, in which case most of the framework goes unused.
About LlamaIndex #
LlamaIndex provides ingestion, indexing, retrieval and query engines for building RAG and agent applications over private data, plus workflow primitives for multi-step agents. The open-source Python framework is MIT licensed; LlamaCloud and LlamaParse are the company's separate hosted, paid services.
LlamaIndex is an open-source project, with 52k stars on GitHub. It was last updated in September 2026.
Make your product part of the discovery. Explore advertising
LlamaIndex vs. the alternatives #
| Record | Stars | Pricing | ||
|---|---|---|---|---|
| LlamaIndexthis listing | 52k | — | MIT | Open source |
| langchain | 146k | Python | MIT | Open source |
| [ruflo](https://agentsearchengine.app/agents/ruflo) | 72k | TypeScript | MIT | Open source |
| [autogen](https://agentsearchengine.app/agents/autogen) | 61k | Python | CC-BY-4.0 | Open source |
| [crewAI](https://agentsearchengine.app/agents/crewai) | 58k | Python | MIT | Open source |
| [smolagents](https://agentsearchengine.app/agents/smolagents) | 29k | — | Apache-2.0 | Open source |
source & further reading
agentsearchengine.app — original article
OpenCode shipped an update
Cline shipped an update
OpenHands shipped an update