Learn about the aims and vision of the WeatherGenerator in this short animated video
The objective of the WeatherGenerator is to improve weather and climate predictions, enhance renewable energy planning, and provide better tools for mitigating the effect of extreme events like floods and heatwaves. It will combine the latest machine learning with high performance computing, creating a robust tool for scientific discovery and real-world applications.
This way, the WeatherGenerator will ensure Europe remains at the forefront of innovation in weather and climate. The resulting European foundation model will be accessible for many diverse applications in the area of Earth system science.
From 27–29 August 2025, project partners of the EU-funded WeatherGenerator project gathered for the first internal hackathon at ECMWF’s facility in Bonn, Germany. This event brought together around 40 researchers, developers, and application partners from across Europe to collaborate in person and lay the groundwork for one of the EU’s most ambitious AI-for-climate projects.
Machine learning powers WeatherGenerator, a global Earth system model for forecasting weather and climate impacts.
Christian Lessig
Running EU-wide Deep Learning projects for Weather The WeatherGenerator project is an ambitious Europe-wide project to build a “Large Language Model for the weather.” From the outset, this leads to a very practical challenge: how to coordinate the development of a state-of-the-art model across 6 High Performance Computing (HPC) systems and 10 research groups? Each HPC system is custom built with the purpose of maximizing processing power at the expense of ease of use or convenience. This experience is very different compared to using a public cloud such as Amazon Web Services, Azure, Google Cloud. Public clouds are meant to be cost effective to build and operate. They are the Ford T’s of large computing: easy to use, simple, adapted to many scenarios. Public cloud operators put a large focus on standardization and automation as well as ease of use for the final customers.
Timothee Hunter,
Christian Lessig
