desiredInferenceUnits
The desired number of inference units to be used by the model using this endpoint. Each inference unit represents of a throughput of 100 characters per second.
The desired number of inference units to be used by the model using this endpoint. Each inference unit represents of a throughput of 100 characters per second.