Create a single duplicate of a model

Select the model in the list.

Right-click the selected model and choose File actions > Duplicate from the menu.
Or
Select the Models > Duplicate model menu option.
Or
Click in the toolbar.

In the Duplicate window, enter the model number in the New model number field.

Change number of the Reference model for the new model, if desired.
By default, this will take the name of the original model selected for duplicating.

Adjust the Model description for the new model.

Check the Change model name in... box to update the model name within sections of the model file itself with the new model number

Check the Use final parameter estimates from reference model box to update parameter estimates in the new model to the ones estimated for the original model.

Check the Make simulation box to make the copied model a simulation model.
When checked, the estimation and covariance steps ($EST and $COV, if present) records are commented out with a semicolon, and a simulation record is inserted into the control file, e.g.,

      $SIM (12345) (54321) ONLY SIM.

Check the Fix all estimates box to use the same estimates in the copied model as in the original.
When checked, $THETA, $OMEGA, and $SIGMA records are modified with the addition of the text “FIX” to fix the parameter estimates, e.g.,

      $THETA
      (0, 1.5,4) FIX ;KA
      (0.1, 0.377,5) FIX ;V/F
      (0, 0.0327,1) FIX ;CL/F 
      (-2, 1.98,2) FIX ;WT on KA

      $OMEGA
      0.281 FIX ; IIV KA
      0.0274 FIX ; IIV V/F
      0.0516 FIX ; IIV CL/F

      $SIGMA
      0.0256 FIX ; properr

Please note that, to correctly duplicate with updated parameter estimates, you are required to adhere to some coding guidelines, especially for the $OMEGA and $SIGMA blocks. See the “NONMEM template control file syntax” for more information.

After pressing the Duplicate button, a new model is created and opened in the editor.

Note:    A useful output file generated when MSF option is used is called INTER, which is an easy read format of the output from NONMEM. Iteration by iteration is listed so the analyst can get ideas for where the run is struggling.