EPT-2 Technical Report

TL;DR

EPT-2 is a new Earth system forecasting AI model that outperforms existing models in predicting key weather variables over 0-240h horizons, with a probabilistic ensemble version surpassing ECMWF standards at lower costs.

Contribution

Introduction of EPT-2, a state-of-the-art Earth physics transformer model, and its probabilistic ensemble EPT-2e, achieving superior accuracy and efficiency in weather forecasting.

Findings

EPT-2 outperforms EPT-1.5 and other AI weather models.

EPT-2e surpasses ECMWF ENS in medium- to long-range forecasts.

EPT models are accessible via the app.jua.ai platform.

Abstract

We present EPT-2, the latest iteration in our Earth Physics Transformer (EPT) family of foundation AI models for Earth system forecasting. EPT-2 delivers substantial improvements over its predecessor, EPT-1.5, and sets a new state of the art in predicting energy-relevant variables-including 10m and 100m wind speed, 2m temperature, and surface solar radiation-across the full 0-240h forecast horizon. It consistently outperforms leading AI weather models such as Microsoft Aurora, as well as the operational numerical forecast system IFS HRES from the European Centre for Medium-Range Weather Forecasts (ECMWF). In parallel, we introduce a perturbation-based ensemble model of EPT-2 for probabilistic forecasting, called EPT-2e. Remarkably, EPT-2e significantly surpasses the ECMWF ENS mean-long considered the gold standard for medium- to longrange forecasting-while operating at a fraction of the computational cost. EPT models, as well as third-party forecasts, are accessible via the app.jua.ai platform.