Create model

The kind of the model provider, can be Provider

The name of the model provider

The arguments to pass to the deployment runtime

The command to run the deployment

The environment variables to set in the deployment. Should be a list of Kubernetes EnvVar types

The Docker image for the deployment

The slug name of the origin model. Only used if the deployment is a ModelDeployment

The readiness probe. Should be a Kubernetes Probe type

The resources for the deployment. Should be a Kubernetes ResourceRequirements type

The type of origin for the deployment

The minimum amount of time that the last replica will remain active AFTER a scale-to-zero decision is made

The maximum number of replicas for the deployment.

Metric watched to make scaling decisions. Can be "cpu" or "memory" or "rps" or "concurrency"

The minimum number of replicas for the deployment. Can be 0 or 1 (in which case the deployment is always running in at least one location).

The time window which must pass at reduced concurrency before a scale-down decision is applied. This can be useful, for example, to keep containers around for a configurable duration to avoid a cold start penalty if new requests come in.

The minimum number of replicas that will be created when the deployment scales up from zero.

The sliding time window over which metrics are averaged to provide the input for scaling decisions

Target value for the watched metric