In this work, a model is built to test the optimal design and performance of a multi-energy systems at five different levels of decentralization, ranging from 225 individual buildings systems to a single district system supplying all building's energy demands. A multi-objective analysis, minimizing both costs and emissions with the epsilon-constraint method, was used to select five Pareto optimal solutions for each level. The considered technologies include gas boilers, oil boilers, photovoltaics, solar thermal panels, ground source heat pumps, air source heat pumps, fuel cells, electrolysers, batteries, thermal storage, and hydrogen storage. It was found that the lowest levels of decentralization (i.e., building or neighbourhood level) were able to achieve the highest level of reduction in carbon emissions and self-sustainability. This is due to the potential for heat pump technologies to be installed on the building level that was not an option on the higher levels of decentralization due to borehole availability and capacity restrictions. On the building level, the combination of PV panels, a heat pump, and a thermal storage tank was found to be the most cost effective and carbon neutral option. (C) 2019 The Authors. Published by Elsevier Ltd.