How to connect to Managed Inference

Configure the CosmicAC CLI and send your first inference request to the Managed Inference service.

Run the interactive setup to configure the CLI with your API key:

cosmicac inference init

Follow the prompts:

Prompt	Description
Protocol	The inference protocol to use (default: `HTTP`)
API key	The API key you created in the prerequisite step
Default model	The model to use for inference requests

The CLI stores your configuration locally. You do not need to provide the key again for subsequent requests.

cosmicac inference chat --message "Explain quantum computing."

A successful response confirms your connection to Managed Inference.