Skip to main content

AI Documents (Elasticsearch)

This module integrates Elasticsearch as an embedding and search provider for the AI Documents feature.

Features​

  • Embedding Support: Uses Elasticsearch for storing vector embeddings of document chunks
  • Search Integration: Retrieves relevant document chunks using Elasticsearch vector search

Getting Started​

  1. Enable the AI Documents (Elasticsearch) feature in Orchard Core admin
  2. Configure an Elasticsearch connection and create an index Search > Indexing. Add a new "Chat Interaction Documents (Elasticsearch)" index.
  3. Select the index in Settings > Chat Interaction

Notes​

The AI Documents feature is provided on demand and will only be enabled when a feature that requires it is enabled (for example, this Elasticsearch provider feature). Ensure you enable this provider feature if you want to configure document indexing with Elasticsearch.