ModelZ CLI is a command line interface that allows you to make predictions with your deployed models.
You could install the CLI by running the following command:
pip install modelz-py
You could use
modelz inference to send a prediction request to the server:
modelz inference -d $DEPLOYMENT_KEY prompt="cute cat"
Or retrieve the deployment metrics:
modelz metrics -d $DEPLOYMENT_KEY
For more details, check the help message and CLI docs (opens in a new tab):
modelz --help modelz inference --help modelz metrics --help
You need to replace
$DEPLOYMENT_KEY with the key of your deployment. Besides this, you need to set the environment variable
MODELZ_API_KEY to your API key. You could get it from the API Keys (opens in a new tab) page.