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.
2025 marked the first year for the WeatherGenerator project, with major advances across all areas - from model architecture to expanded datasets and growing community engagement. Let’s have a look at some of the project’s highlights of the year.
Join Us in Oslo – WeatherGenerator Dissemination Workshop 2026! We are excited to announce the upcoming WeatherGenerator Dissemination Workshop for everyone working with climate data, environmental modelling, and weather-related decision-making. 📍 Location: Hybrid - Online and in Oslo, Norway 📅 Date: 11 February 2026 The day event will open with a summary of the project, followed by detailed discussion with the project partners on the WeatherGenerator achievements. Invited guest speakers will highlight the relevance and usage of the WeatherGenerator and its tools / applications, and we will also have highlight presentation from related international efforts in the area of foundation modelling for weather and climate science.
Explore the New WeatherGenerator Theme Videos We are excited to introduce a brand-new collection of videos that takes you behind the scenes of the WeatherGenerator project. Four Theme Videos – each dedicated to one core pillar of the initiative – along with an overall summary video are now available on our website. Each Theme Video highlights a key element of the project: Theme 1: From Raw Data to Ready-to-Use Inputs This video explores how the project brings together diverse climate and weather datasets, transforming them into a unified, scalable foundation tailored for machine-learning workflows.
Video Highlights from Our First Hackathon We are pleased to share highlights from WeatherGenerator’s first internal hackathon, held from 27–29 August 2025 in Bonn, Germany at ECMWF. Over the course of three days, around 40 researchers, developers, and application partners from across Europe came together to collaborate, exchange knowledge, and lay the foundation for one of the EU’s most ambitious AI-for-climate projects. From data foundations to real-world applications, the hackathon was not only about technical progress, but also about building connections, sharing insights, and shaping the future of climate intelligence.
WeatherGenerator’s Sister Projects WeatherGenerator is one of three projects launched in January and February 2025 under Horizon Europe to strengthen Destination Earth (DestinE), Europe’s initiative to build a high-precision, interactive digital twin of the Earth. Digital twins are computational models of real-world phenomena that can be used to simulate the past, model the present, and predict the future. Together with TerraDT and UrbanAIR, we expand the DestinE ecosystem by addressing different but complementary aspects of the Earth system.
What are Supercomputers? Supercomputers are high-performance computing systems designed to solve problems which exceed the capacity of conventional machines. Their defining feature is massive parallelism: instead of relying on a single processor, they connect tens of thousands of CPUs and GPUs which perform calculations simultaneously. This parallel architecture allows them to reach speeds measured in petaflops (10¹⁵ operations per second) or even exaflops (10¹⁸ operations per second), enabling computations that would take years on a standard computer to be completed within hours or days.
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.
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
Machine learning powers WeatherGenerator, a global Earth system model for forecasting weather and climate impacts.
Christian Lessig
