Lumen
An open-source, extensible agent framework for chatting with data, turning natural language into SQL, charts, dashboards and reports.
Lumen is an open-source, extensible agent framework for chatting with data and for retrieval-augmented generation over documents. Its declarative data model lets an LLM generate entire data transformation pipelines and visualizations, which can then be serialized, shared, continued in a notebook, or assembled into a dashboard.
It generates SQL over local or remote files, SQL databases or a data lake, can be given documents and embeddings for added context, and turns natural-language requests into charts, tables and dashboards. Every LLM output can be inspected, refined and manually edited, and the LLM can summarize results and key insights. It can be extended with custom agents, tools and analyses for a particular domain.
It is built in Python on top of Panel, which lets it render almost any type of output, and focuses on staying fully open and extensible rather than being a closed pipeline.