The wind energy industry widely recognizes that wind flow modeling remains the largest individual source of uncertainty in most energy production estimates. Nevertheless, there has been great progress since the first seminal linear wind flow models in the 1970s. Given the exponential increase in computing resources over the last few decades, more advanced numerical models based on mesoscale numerical weather prediction (NWP) models and/or computational fluid dynamics (CFD) models are now the norm with Large Eddy Simulations (LES) starting to gain traction within the wind industry. In addition, machine learning tools are finding applications in wind resource assessment. Despite these advances numerical models will never be perfect at least not until direct numerical simulations (DNS) are possible. Therefore, understanding their weakness and quantifying their uncertainty should be a priority. This session will cover current and next generation wind flow models and will inform about their accuracy and uncertainty.