All open source
Open SourceRAG2k

SWIRL AI Search

Federated RAG search across all your apps with no data movement.

SWIRL is an open-source AI search platform that runs federated, retrieval-augmented queries across many enterprise data sources and APIs at once, then reranks and summarizes results with an LLM. It queries systems in place rather than copying data into a central index, with connectors for common SaaS apps, databases, and document stores. It is notable for doing RAG over distributed sources without building a unified data lake first.

Repository

swirlai/swirl-search

Language

Python

What you'd build with it

  • Running a single AI query across multiple apps and document stores without first ETLing everything into one index
  • Building enterprise search that retrieves, reranks, and summarizes results from heterogeneous sources
  • Prototyping RAG over existing systems while respecting source-level access controls

Tags

federated-searchragenterprise-searchretrieval