As the main consumers of urban energy, the low carbon and efficient operation of intelligent buildings is crucial for achieving the goals of “peak carbon dioxide emissions and carbon neutrality”. To enhance building economics while improving energy sharing and the consumption of distributed energy resources, a distributed optimal scheduling model for buildings taking into account building characteristics and energy trading was proposed. At the building optimization level, a multi-objective building operation optimization model considering economic and temperature comfort requirements was established. At the energy sharing level, a peer-to-peer buildings trading market was set up, and a new continuous double auction trading mechanism combining building optimization results and market risks was proposed. This mechanism feeds back the results of market transactions to the optimization level of each building, achieving iterative optimization and energy sharing within buildings. Robust optimization was used to test the effectiveness of the model in various uncertainty scenarios. Simulation results show that the distributed optimal scheduling model can optimize building economics while enhancing energy complementarity and the consumption of distributed energy resources in various scenarios.