.svg)
Rocket.Chat AI (RAI) is a privacy-first, enterprise framework that provides AI capabilities for your Rocket.Chat workspace, while giving you control over how your data is accessed, processed, and used. RAI includes AI-powered search and conversational capabilities built on Retrieval-Augmented Generation (RAG), as well as integrations that allow external AI clients to interact with Rocket.Chat.
Administrators can manage Rocket.Chat AI from Manage 
> Workspace > AI Center.
Key capabilities
AI Search
A hybrid search engine that combines semantic, intent-based search with traditional keyword search to retrieve relevant information from your workspace based on access levels. When an LLM provider is connected, it can also generate concise natural language answers from those search results. See the official AI Search documentation to get started.MCP Server
A standardized interface that exposes Rocket.Chat tools and context to external AI clients, allowing them to securely interact with your workspace. Refer to the MCP Server guide for more details.Rocket.Chat AI App
An AI assistant for workspace conversations. Users can interact with the AI bot directly or mention it in a room to ask questions about indexed organizational knowledge. The bot can also summarize conversations and threads, highlighting key discussion points and action items. Visit the Rocket.Chat AI app documentation to learn more.
Both AI Search and Rocket.Chat AI App require the self-hosted Rocket.Chat AI pipeline for RAG to retrieve and process indexed workspace data. Visit the Deploy Rocket.Chat AI Pipeline guide to learn more.
Next steps
Deploy the AI Pipeline to configure the infrastructure required for RAG-based AI capabilities like AI Search and Rocket.Chat AI App.
Configure AI Search to combine keyword and semantic intelligent search in your workspace global search bar.
Set up the Rocket.Chat AI app to use an AI assistant in your workspace.
Configure the MCP Server to connect external AI clients to Rocket.Chat.