The plan is to support operator overloading to implement Int + Float and others, in order to get the desired behavior at an acceptable performance level.

Until then, I recommend not using explicitly the Numeric static type if you don't need it, the Numeric limitations should not affect the direct use of Float and Int. You probably already noticed, in the given example simply removing the two :Numeric types will result in the expected static error to be displayed. But if you do need to use Numeric, use it with the Numeric-safe (but slower) methods add, sub, etc.