This paper discusses the need for critically evaluating regional-scale (~200–2,000 km) three-dimensional numerical photochemical air quality modeling systems to establish a model’s credibility in simulating the spatio-temporal features embedded in the observations. Because of limitations of currently used approaches for evaluating regional air quality models, a framework for model evaluation is introduced here for determining the suitability of a modeling system for a given application, distinguishing the performance between different models through confidence-testing of model results, guiding model development and analyzing the impacts of regulatory policy options. The framework identifies operational, diagnostic, dynamic, and probabilistic types of model evaluation. Operational evaluation techniques include statistical and graphical analyses aimed at determining whether model estimates are in agreement with the observations in an overall sense. Diagnostic evaluation focuses on process-oriented analyses to determine whether the individual processes and components of the model system are working correctly, both independently and in combination. Dynamic evaluation assesses the ability of the air quality model to simulate changes in air quality stemming from changes in source emissions and/or meteorology, the principal forces that drive the air quality model. Probabilistic evaluation attempts to assess the confidence that can be placed in model predictions using techniques such as ensemble modeling and Bayesian model averaging. The advantages of these types of model evaluation approaches are discussed in this paper.