Aurora is a machine learning model that can predict atmospheric variables, such as temperature. It is a foundation model, which means that it was first generally trained on a lot of data and then can be adapted to specialized atmospheric forecasting tasks with relatively little data. We provide four such specialized versions: one for medium-resolution weather prediction, one for high-resolution weather prediction, one for air pollution prediction, and one for ocean wave prediction.
MS created foundation model Aurora, trained on over '1 million hours of diverse geophysical data' (they mean 1 million compute hours??), to use to predict atmospheric variables (temp) in August 2024.