I assume that you mean a bivariate regression. E.g. height vs weight.
I think you will have to calculate the tolerance interval.
You should be able to get close by using Distribution to calculate the tolerance interval of the residuals.
In the attached example I have fit height vs weight (Fit Y by X) and saved the residuals. Then used Distribution to calculate the prediction interval on the residuals. You can see that these calculated limits (purple and orange) are close to the prediction interval from Fit Y by X, "...95% Indiv height" (red and green).
I think you should be able to do the same for tolerance interval.
Big Class with Bivariate prediction interval calculations.jmp