Allows to quantify tensors, downcast values, process in float16 etc.
@Int32() external int allowPrecisionLoss;