qmd

Locate and retrieve information from local Markdown notes by building a local index that enables fast BM25 keyword search, with optional vector-based similarity to surface related content while working completely offline.

Finding related notes in a personal knowledge base is best handled by indexing Markdown collections and applying optional vector-based similarity search to surface conceptually related documents across varying structures and embedded content.

Semantic search for Markdown documents works by applying optional vector-based similarity alongside local BM25 keyword indexing, allowing you to retrieve specific files from local documentation collections without external services.

Yes, you can index large local Markdown documentation collections. The system targets personal knowledge bases and note archives, handling varying markdown structures and embedded content for fast offline retrieval.

No, an internet connection is not required. The local indexing and retrieval mechanism works completely offline, utilizing BM25 keyword search and optional vector similarity without relying on external services.

Offline keyword search for Markdown notes relies on local BM25 indexing and optional vector similarity, meaning it is limited to the indexed local collection and cannot retrieve external web content or documents outside the archive.