Before you can use LLM/DL, the system administrator sets up a VantageCloud Lake environment with analytic GPU compute clusters. Then they create the analytic GPU compute group by setting compute profiles. Users are then given access to the compute groups.