Multifamily Building

One clarification might be that although the A/C unit may be located in the living room, the bedroom may still be cooled from that unit either indirectly or by allowing a duct between the rooms or transfer air.

What is the location (general description if you can't be specific) and maybe if you can describe the heating system we'll have a better feel for your building. 

If the whole facility is heated, and the bedrooms might be indirectly cooled, I would still model the whole building as being conditioned but perhaps the bedrooms may have a higher temperature setpoint during the summer season. 

The model may over estimate energy consumption then if some occupants shut their doors and prevent the bedrooms from being conditioned or if the occupants may tend to use operable windows for cooling.

It sounds like the bedrooms are indirectly cooled, so they would be part of the HVAC zone that includes the entire apartment. The location of the unit does not really matter. The entire space is conditioned, so it needs to have a system for baseline. The system in the baseline will be sized for the whole zone. You might be able to justify the creation of a separate zone in the bedrooms, with a different temperature setting but I would think this would need to be backed up with some considerable study of how the temperatures would vary within the actual design.

How are the air conditioning units being sized?

The nature of the partitions and the door opening will influence whether this is indirectly conditioned or not. This comes down to whether the bedroom spaces meet the definition of indirectly conditioned. So the first thing I would do is to look at the definition and make that determination. If it is not indirectly conditioned then it is unconditioned. If it is unconditioned then it will be unconditioned in both models. If it is indirectly conditioned and the unit is only sized for the living space you may run into unmet load hour issues. If you do you will probably have to increase the temperature settings until you get the unmet load hours below the limit. The temperature setting must be identical in both models.

As an aside if the whole unit is fully conditioned in the baseline and is not in the proposed this would not be fair comparison but it would increase the baseline energy use and therefore, increase your savings. So I agree that both models should include the bedrooms as indirectly conditioned or as unconditioned in order to generate a fair comparison.