RAG retrieval demos
Retrieval is the first half of RAG: split text into chunks, turn each chunk into a vector, and find the ones closest in meaning to a question. These two pages let you watch that happen. Nothing is uploaded; your files stay in your browser.
Demo 1
Document demo
Drop in a single file, watch it get chunked by size or by meaning, then ask a question and compare semantic search with keyword search.
- .txt, .md, .csv and .pdf
- Fixed-size and semantic chunking
- Semantic vs keyword side by side
Obsidian vault demo
Drop in a vault folder (or try the built-in sample). Notes are split along their own headings, and every result tells you which note and section it came from.
- Folder drop, frontmatter and wikilinks handled
- Chunk by heading vs fixed size
- Results labelled with note and heading path