// the find
run-llama/LlamaIndexTS
Data framework for your LLM applications. Focus on server side solution
TypeScript/JS port of LlamaIndex for building RAG and agent apps against your own data, with support for Node, Deno, Bun, and edge runtimes like Cloudflare Workers and Vercel Edge. It targets JS/TS developers who want the LlamaIndex data-framework pattern without switching to Python.
Provider packages are split out from the core (@llamaindex/openai, etc.), so you only pull in what you use instead of one monolithic dependency. Multi-runtime support including edge runtimes is a real differentiator versus most Python-first RAG tooling. The docs site has actual migration guides (0.8 to 0.9) and deprecated-feature notes, which is more than most JS LLM libraries bother with.
The repo is archived/deprecated as of this snapshot - the README opens with a deprecation notice and points only to Python LlamaCloud/LlamaParse docs, with no stated TS successor or migration path. Browser support is explicitly broken because of missing AsyncLocalStorage-like APIs, so the 'multi-runtime' story has a real gap. Starting a new project on a framework whose maintainers have already walked away means you're on your own for bug fixes and provider updates (new model releases, breaking API changes) going forward.