Alberto asked me for an opinion about this Litobox device. So below, please find a few points for inspiration, mostly stuff to ask the Litobox people about to inform your decision. I am not a heating system technician, so take it with a lot of salt.
Overall I like their “heat pump as appliance” idea, as it is more DIY and could easily be more reliable and perhaps also cheaper. The question is if this system is the most suitable to provide what you want, esp. in combination with self-generated electricity:
Agreed, but an issue with demand planning might arise since heat pumps are easily the most power hungry energy takers of the house. It gets interesting in spring and autumn when sometimes there is more PV supply than energy demand, sometimes the opposite, so by buffering and planning demand in a collectively coordinated way the limited PV supply can go farther. How to do that collective demand planning I don’t know. But the heat pump offers one more tool for that: the hot water tank (integrated into the Litobox as per the article) can buffer energy in the form of hot water. More cost effective than batteries, and very useful together with the weather forecast (both temperature and insolation for PV generation) and a dynamic electricity rate.
This also brings me to the main caveat I have about the Litobox: how large is its buffer tank? It does not say in the article, but it might be too small to allow for any significant savings by generating hot water in advance during times when electricity is cheap (and thus also contributing to grid stability by adding energy storage). For our two-family home I would add ca. 1500 l of hot water buffer tank. It does not seem that, for eight units, there are 4 × 1500 l = 6 m³ of hot water buffer tanks in the Litobox. It would also not be feasible due to roof loading.
So I think you can do a simulation to know how much hot water buffer tanks you would like to have for saving on heating costs (and potentially cold water in the same tanks, saving on cooling costs …). And then ask the Litobox people in which way this can be integrated into their system. The piping is not a problem, as a hot water buffer tank acts like another heat demand / radiator. The question is if the Litobox control system can be configured to charge a buffer tank, or if you need your own digital control system for that. (See note below on what Litobox should offer for external digital control.)
It has both PV panels and thermal solar collectors? That would be a rather strange system design, would be good to ask them about.
My argument against integrated PV panels: PV panels on the box itself would never be enough to power the heat pump in blackout situations, so why add them at all? To get off-grid capabilities people would need to add a large PV array around the box anyway, and use regular grid inverters, batteries etc.. When doing so, they would want to avoid having the same kind of system (at maybe 5% its power output) inside their heat pump, as that just adds complexity and points of failure but not functionality.
My argument against integrated thermal solar collectors: this whole class of equipment has no justification once one decided to have a heat pump and PV panels already. Because these have a higher heat generation efficiency per collector area (very roughly 22% for PV panels times COP 4 for a heat pump = 88%), and are more versatile by making electricity in summer instead, or even delivering cooling power via the same heat pump. So thermal panels would only add system complexity and points of failure without adding function.
In my opinion (software engineer here), it really depends how they implement and integrate their digital interfaces and functions. If it’s the equivalent of a washing machine where the door does not open without a wifi connection, then I’d not touch this with a 10 ft pole of course. But to future proof for the necessary energy transition, with demand flexibility and all, there is really no way around having digital control over all large electricity consumers in the house. So there has to be a way to control the heat pump (operation, power level, output temperature) and sense its environment (primary circuit input and output temperature, buffer temperature, heating circuit forward and return temperature, heat demand) via a digital protocol. If that protocol is relatively widespread, can use or be adapted to Ethernet cable, and has open source implementations, all is good. You can then integrate it somehow into your digital building control environment. Worst case, it may need a microcontroller and some custom programming (not such a big deal anymore, AI will help 
).
In the article linked above, they mention:
One large heat pump for multiple apartments is much quieter than a roof full of individual outdoor units.
So that might be a good argument to tell the neighbors.