In the model, the connection is not rigid neither flexible, so the beam pass some value of moment to the column. A beam with flexible connection in both extreme, the moment will be zero in the extremes, and maximum in the middle, with the value M = (q*L^)/8.

Maybe I have to consider rigid connection and compare this moment, baecause the article compare a cantilever beam and compare the full moment. I am thinking: when this moment is not transfered to the column, the beam rotate.

@Sandeep, in the article we have reactions with cross section models. 

It says: "For a Moment Reaction scoped to a contact region, the location of the summation point may not be exactly on the contact region itself."

In my case, it is exactly what is happening. I would like to know the moment transfered between the contact region: bolts and endplate, endplate and column, and bolts and column.