Core Concepts

Vector Search

Semantic search that understands meaning, not just keywords.

Vector search uses AI embeddings to find content based on semantic similarity. This means searching for "pricing" will also find content about "costs," "fees," and "subscription plans."

How It Works

  1. 1

    Text to Vector

    Your content is converted into high-dimensional vectors (embeddings)

  2. 2

    Query Embedding

    User questions are also converted to the same vector space

  3. 3

    Similarity Search

    Find content vectors closest to the query vector

Technology Stack

ChromaDB

Purpose-built vector database for fast similarity search

OpenAI Embeddings

State-of-the-art text embeddings for semantic understanding

Feedback

Route: /docs/vector-search