QMD is an on-device search engine designed to index and retrieve knowledge from markdown notes, documentation, and transcripts using a hybrid approach. It combines keyword search (BM25), vector semantic search, and LLM-based re-ranking into a unified pipeline that runs locally. The project includes a fine-tuning workflow to train custom query expansion models and exposes its capabilities to external AI agents via the Model Context Protocol (MCP).
qmd is organized as connected concepts and components. Start broad, then drill down chapter by chapter.
QMD is an on-device search engine designed to index and retrieve knowledge from markdown notes, documentation, and transcripts using a hybrid approach. It combines keyword search (BM25), vector semantic search, and LLM-based re-ranking into a unified pipeline that runs locally. The project includes a fine-tuning workflow to train custom query expansion models and exposes its capabilities to external AI agents via the Model Context Protocol (MCP).
Source Repository: https://github.com/tobi/qmd
Follow sequentially or jump to any topic. Start with Hybrid Search Orchestrator.
This tutorial was automatically generated by Code IQ and rendered with the shared tutorial site builder. It can be produced for any repository tutorial folder that follows the numbered markdown chapter layout.
View Code IQ ↗