Electricity production is the sector with the largest share of global emissions and there are many options for decarbonizing it. Identifying the lowest cost option for achieving decarbonization (and full reliability) is a complex optimization problem at the intersection of economics and engineering. Key determinants are the cost of individual technologies, the geographical potential, the complementarities between energy sources and supporting infrastructure like electricity grids and energy storage. This paper reviews the literature on the subject and draws high-level conclusions from the abundance of specialized analyses. It finds that energy-economy models have strongly changed projections of the optimal electricity mix in recent years. While the models differ in detail, models project that the share of renewable energy, mostly solar and wind power, increases steadily in a “below 2°C” scenario and becomes the dominant source of energy by 2050. An electricity system based on solar and wind power can use flexibility options as a complement instead of baseload energy. Models vary by the degree to which renewable energy is supported by carbon capture and storage, bioenergy, and nuclear energy.