04/Knowledge (RAG)
Playground & AI Providers
Testing knowledge and configuring the LLM/embedding model.
The Playground lets you try queries against the generated MCP/RAG and inspect retrieved chunks and the answer. AI Provider settings determine the LLM and embedding model (e.g. nomic-embed-text) used by the RAG pipeline.
The embedding model is also used by the gateway's Semantic router and by Second Brain.