Show HN: OpenFable – Open-source RAG engine using tree-structured indexes
Category: ai-ml
Tags: rag, retrieval-augmented-generation, semantic-search, llm, vector-database
Score: 7.0/10 (Innovation: 7, Technical: 7, Documentation: 8, Utility: 6)
OpenFable is an open-source retrieval engine that implements the FABLE algorithm, building LLM-enhanced semantic forest indexes from documents for structured, budget-aware retrieval. It's interesting because it moves beyond flat vector search by using a tree hierarchy and bi-path retrieval (LLM reasoning + vector similarity) to better handle complex queries spanning multiple document sections.
Target audience: data engineers, backend devs, ai researchers
Repository: https://github.com/alainbrown/openfable · Python · Apache-2.0 · 3 stars
View on Hacker News