Embeddings

Start with a few plain-text documents. Split them into passages, generate an embedding for each passage with Ollama, and embed each search query with the same model. Rank passages by…