Advancing Earth system science with innovative AI

Project Overview

Learn about the aims and vision of the WeatherGenerator in this short animated video

Groundbreaking European initiative aimed at transforming Earth system science through advanced AI and high-performance computing.

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.

Project goals

European AI model built for weather and climate
Enhance predictions across timescales
Deepen Earth system understanding
Optimize energy planning and disaster aid
Offer open-source tools and data
Train AI experts in Earth sciences

Research highlights & breakthroughs

- #community
- #Mentorship
1 min read • Apr 13, 2026
Introducing the WeatherGenerator Mentorship Program
Whether you are early in your career or an experienced researcher, meaningful connections can shape the path ahead. The WeatherGenerator Mentorship Program brings together professionals and researchers across career stages in an informal setting built for honest exchange and growth. The WeatherGenerator Mentorship Program is a 6-month initiative connecting experienced professionals and researchers with early career peers. Mentors share their experiences and perspectives, contributing to a diverse community, while mentees gain space to reflect on their goals, explore career topics, and grow their network.
- #Science Explainer
1 min read • Apr 02, 2026
Science Explainer 1: Can weather models predict power prices?
In the first instalment of the WeatherGenerator Science Explainer series, Even Nordhagen from Norwegian Meteorological Institute explains how the WeatherGenerator goes beyond traditional weather models, directly predicting energy - relevant variables like wind power output and reservoir inflow, without the need for separate downstream models. With a built-in long-term memory, the system can track slow-building processes like snowpack and soil moisture, and ultimately learn the full chain from atmospheric conditions to electricity markets.
1 min read • Mar 26, 2026
Video Highlights from Oslo
From 10–12 February 2026 WeatherGenerator project partners and participants met at Forskningsparken – Oslo Science Park, as well as online, for a dynamic programme combining an internal General Assembly, a public Dissemination & Stakeholder Day, and a hands-on Coding Workshop. The programme covered the technical foundations of WeatherGenerator: its architecture, training approach, data formats and applications - alongside contributions from our sister projects TerraDT and UrbanAIR and invited speakers working on AI-driven approaches to weather and climate science, including foundation modelling.
- #community
- #article
5 min read • Mar 05, 2026
Women in WeatherGenerator: Celebrating International Women's Day
Science advances through the courage to ask bold questions and across the WeatherGenerator project, women are at the forefront of asking them. From developing the core AI architecture to adapting it for real-world applications in energy, health, and extreme weather forecasting, female researchers and engineers are shaping the future of AI-driven meteorology every day. WeatherGenerator is an EU Horizon project bringing together 16 leading European organisations to build a next-generation foundation model for weather and climate. Behind this mission is a diverse community of scientists, and on this International Women’s Day, we want to shine a light on some of the extraordinary women who make this project what it is. Coming from across the globe, bringing expertise in atmospheric chemistry, machine learning, probabilistic modelling, remote sensing, and operational meteorology.
1 min read • Feb 17, 2026
WeatherGenerator Meets in Oslo and Online
One year into the project, partners came together for three days of exchange, discussion and hands-on collaboration in Oslo, Norway. It was great to meet in person, connect with new partners, and strengthen existing collaborations. From 10–12 February 2026, project partners and participants met at Forskningsparken - Oslo Science Park - and online for the WeatherGenerator event hosted by Norwegian Meteorological Institute | Meteorologisk institutt. The programme combined an internal General Assembly, a public Dissemination & Stakeholder Day, and a focused Coding Workshop.
- #workshop
1 min read • Jan 22, 2026
Advancing AI Climate Models - AlgoEarth & CESOC Workshop
We would like to invite you to the AlgoEarth and CESOC Workshop “Advancing AI Climate Models” Our goal is to connect expertise across algorithm design, machine learning, and Earth system science, fostering interdisciplinary thinking around next-generation AI methods designed for emerging global observation platforms and their use in weather and climate prediction. The event will take place from Wednesday, 04 March 2026 (12:00 CET), to Thursday, 05 March 2026 (13:00 CET) at the University of Cologne, Pohligstraße 3, 50969 Cologne, Germany, in lecture hall 4.001.

Explore news

Technical Blog Vol. 2
Orchestrating Deep Learning Across Europe's HPC Landscape
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
Aug 05, 2025 6 min read
Technical Blog Vol. 1
The WeatherGenerator: building a European foundation model for Weather and Climate
Machine learning powers WeatherGenerator, a global Earth system model for forecasting weather and climate impacts.
Christian Lessig
May 07, 2025 3 min read