Long-term energy scenarios are an important input to policy-relevant assessment reports on climate change such as those produced by the Intergovernmental Panel on Climate Change or the United Nations Environment Programme to just name a few examples. They are also used by government agencies to support their decision making in the context of climate change mitigation and other energyrelated challenges. In response to this demand, two broader categories of model development are currently pursued by the scientific community: (1) the degree of integration is increasing, in other words, the system boundaries of models are being extended, in particular to address the interlinkages between the energy, land, food, water, and climate more comprehensively and (2) the heterogeneity of the representation of various entities (e.g., spatial, sectoral, socio-economic) is increasing to adequately address distributional effects (e.g., countries within regions, urban vs rural areas, different types of households). Moreover, the energyclimate scenarios that are being developed are designed to be more 'realistic' by going beyond very stylized designs and integrate features that are observed in the real world. This includes delayed action on climate mitigation or fragmented approaches to mitigation that not only exclude major emitters from climate action, but also the exclusion of specific technologies from the portfolio of mitigation options in response to technical challenges or public acceptance issues. Finally, an attempt is made to summarize robust insights that have emerged from individual studies and particularly from modeling comparison exercises.as to government agencies directly. The majority of the existing energy scenario literature relies on formal quantitative models, although more qualitative approaches are also used. 2,5 Quantitative models are widely different in terms of their system boundaries, i.e., which natural, human and technical systems they cover, the detail of representation of various system parts and the (mathematical) solution concept. As a result, models have different strengths and weaknesses that are important to take into account when interpreting their results. 6 Modelers are typically well aware of the limitations of their models and results whereas users of model results are sometimes not-not only because of insufficient communication from the modelers, but also because of misunderstandings about the purpose of model-based scenario analyses (see Box 1) and lacking expertise of decision makers about the models and their results.