Yes, that geometry is WAY too complex for an energy model. EnergyPlus does not do 3D modeling of heat flow in either the ground or the zone, so there is no use trying to maintain real-world geometry, and some calculations grow as a power law or even exponentially with the number of surfaces. Start simple, say with an octagonal shape and maybe 2-3 steps: you will still have the same approximate volume and ground wall/floor surfaces, and similar radiation exchange.
You should be aware that the default behavior for ground coupling can be very inaccurate; see the discussion in this thread or the EnergyPlus documentation here. For such a large, deep below-ground volume, you might be able to get a reasonable result following the EnergyPlus documentation’s recommendation: “for typical commercial buildings in the USA, a reasonable default value is 2C less than the average indoor space temperature.” Again, note EnergyPlus provides a lot of cautions on using that.