The number of inference component copies currently placed on instances of this type.
The ML compute instance type where the inference component copies are placed.