But you're right, they have an OpenAI compatible API https://inference-docs.cerebras.ai/resources/openai so perhaps I can actually use this in the CLI! Thanks for making me take another look.
EDIT: Woah, Charm supports this natively. This is great. I am going to try this now.