Rocket.Chat AI app FAQs
Do users need to bring their own environment components, such as LLMs?
Yes, users must bring their own LLM deployment. Rocket.Chat does not provide any hosted options. You can also use Cloud LLM inference providers.
What are the requirements for LLMs that can be used with Rocket.Chat?
The RAI app can connect with any LLM as long as the deployment is an "OpenAI compatible endpoint." For example, you can use Claude or Gemma.
Is the Milvus vector database required?
Milvus is used for RAG capabilities, and it's not mandatory. Additionally, Rocket.Chat has a dedicated RAI pipeline that handles everything out of the box, with additional features (like encryption and RBAC on data), and can also be used for the Intelligent Search app. If no RAG is configured, the AI app can summarize chats and discussions and run an LLM-only chat in the DM.
Can I search chats with the AI app?
You can search using the Intelligent Search app.