Introduction

This is the user guide for machine learning (ML) services at ECMWF.

Further support is available through the following channels:

  • For enquiries relating to data access, computing resources and related issues, please raise a ticket via the ECMWF Support Portal.
  • To follow announcements and engage in discussions with the user community and ECMWF staff, visit our User Forum.
  • To raise a bug report or contribute to Anemoi developments, please create a Github issue in line with our contribution guidelines


News Feed

Users may 'watch' this page to receive notifications of updates and follow our AIFS blog for the latest developments.


User Documentation

AIFS 

OPERATIONAL

The AIFS is ECMWF's Artificial Intelligence Forecasting System. First launched in October 2023, the AIFS generates deterministic and ensemble weather forecasts from machine learning models. 

Details about the current operational versions of AIFS models, and a log of past model versions, can be found on AIFS Version History.

Meteorological information about the AIFS, including a guide for forecasters, is documented in the Forecast User Guide: Section 2.1.6 Machine Learning models

How-to guides

FAQs

AIFS News


Anemoi

INCUBATING (see ECMWF's guidelines on software maturity)

Anemoi is an award-winning open-source, Python-based framework developed collaboratively by ECMWF and several European national meteorological services. It is designed to facilitate the development, training, and deployment of machine learning models for weather forecasting. As an 'end to end' framework, it provides a comprehensive toolkit that spans data preparation, model training, and inference, enabling meteorological organizations to leverage their own data for ML-based weather prediction. Anemoi packages are listed below, together with their dedicated user documentation. 

In January 2025, a webinar series took place exploring various components of the Anemoi framework. Webinar recordings and presentation materials are freely available on the ECMWF website.

Anemoi componentDescriptionDocumentationWebinar link
anemoi-core

A mono-repo containing core training and modelling functionality for Anemoi. Packages include:

anemoi-graphs: Provides the functionality to create complex global or local area graphs. 

anemoi-models: Provides implementations for various type of models. These models are based on a graph encoder-processor-decoder approach and are implemented using the PyTorch library. 

anemoi-training: Provides the functionality to train machine learning models, using pytorch-lightning and Hydra. The training is highly configurable and fully defined through configuration files (utilising anemoi-models to achieve this). The package also includes profiling evaluation, plotting and logging of defined model and system metrics. anemoi-training is designed to work with datasets created using anemoi-datasets.

https://anemoi.readthedocs.io/projects/graphs/en/latest/

https://anemoi.readthedocs.io/projects/models/en/latest/


https://anemoi.readthedocs.io/projects/training/en/latest/

21 January 2025

23 January 2025

anemoi-datasetsProvides the tools to build datasets which are optimised for machine learning training, with appropriate chunking and precomputed statistics for normalisation. These datasets can be built from a range of input sources, including MARS, Grib, NetCDF, Zarr and more. https://anemoi.readthedocs.io/projects/datasets/en/latest/15 January 2025
anemoi-inference

Provides the tools to take a trained model and perform the inference/rollout given some initial conditions. Inference makes full use of the metadata stored in a checkpoint to facilitate simple execution without requiring large amounts of boilerplate code.

https://anemoi.readthedocs.io/projects/inference/en/latest/28 January 2025
anemoi-registry

Provides the tools to save a dataset, a model or an experiment to the Anemoi catalogue so that it can be easily shared with others. The catalogue is accessible for permitted users only.

https://anemoi.readthedocs.io/projects/registry/en/latest/
anemoi-transformContains data transformation functions which can be applied to datasets (via filters).https://anemoi.readthedocs.io/projects/transform/en/latest/
anemoi-utilsContains miscellaneous utility functions which are used across the other packages.https://anemoi.readthedocs.io/projects/utils/en/latest/

FAQs

Anemoi News


ai-models

SANDBOX (see ECMWF's guidelines on software maturity)

ECMWF runs several third-party data-driven weather forecasting models alongside the AIFS. These state-of-the-art models are developed externally and are not maintained by ECMWF. The models are run daily and graphical output is publicly available on the OpenCharts platform. Raw data are not available to users.

Users can also generate their own forecasts from these models using the experimental ai-models package. Please note that support for third-party models accessed through ai-models is provided on a best-efforts basis, and may not always be possible. For model-specific functionality, we recommend contacting the original model developers.

Please note that, as ECMWF is co-developing the Anemoi framework for operational deployment of machine learning models, the ai-models interface will remain experimental, with no current plans for expanding its features.

How-to guides

FAQs

ai-models News


GPUs

GPUs are available through the ATOS high performance computing facility (HPCF) and the European Weather Cloud. GPU usage is permitted only got approved users affiliated with National Meteorological and Hydrological Services in the Member and Co-operating States of ECMWF. GPU access may also be made available for research in the context of ECMWF Special Projects.

See Access to Computing Facilities for further details on how to request access to GPU resources.

Technical GPU documentation can be accessed via the following links:

GPUs News


Further Resources

DateTypeLink
March 2026

TRAINING

"Data assimilation & Machine Learning"

February 2026

TRAINING

"Machine learning and Destination Earth"

November 2025

TRAINING

"Machine Learning for operational forecasters"

October 2025

MEDIA

"Introducing the Anemoi training-ready version of ERA5"

October 2025

TRAINING

"Machine learning for weather prediction"

August 2025

MEDIA

"Representing ocean wind waves in ECMWF's AIFS"

July 2025

MEDIA

"Unlocking the black box: the potential of explainable AI in geoscience"

July 2025

MEDIA

"ECMWF’s ensemble AI forecasts become operational"

June 2025

WEBINAR

"Introduction to AIFS ENS v1" (video)

May 2025

MEDIA

"Verifying 2 m temperature forecasts in wintertime anticyclonic conditions"

May 2025

MEDIA

"The AI Weather Quest Testing Period is live"

May 2025

MEDIA

"Florence Rabier and Anemoi to receive European Meteorological Society awards"

April 2025

MEDIA

"Scientists present new ML tool for improved fire prediction"

April 2025

MEDIA

"Operational release of AIFS Single 1.0"

March 2025

TRAINING

"Data assimilation & Machine Learning"

February 2025

WEBINAR

"Introduction to AIFS Single v1" (video)

February 2025

MEDIA

"ECMWF's AI forecasts become operational"

February 2025

TRAINING

"Discover Anemoi: Kicking off our 2025 machine learning training"

January 2025

MEDIA

"An update on AI–DOP: skilful weather forecasts produced directly from observations"

December 2024

MEDIA

"A year in ML for weather forecasting"

December 2024

MEDIA

The AI revolution: how European weather services are harnessing innovation

September 2024

MEDIA

"Machine learning to play growing role in weather forecasting, says DG"

March 2024

TRAINING

"Machine learning for weather prediction"

January 2024

MEDIA

"Red sky at night... producing weather forecasts directly from observations"

July 2023

MEDIA 

"All eyes on high-performance computing in meteorology"

June 2023

MEDIA

"The rise of machine learning in weather forecasting"

January 2023

TRAINING

Massive Open Online Course (MOOC): Machine Learning in Weather & Climate

2022

MEDIA

ECMWF Annual report - "Machine learning in numerical weather prediction"

April 2021

MEDIA

"Data assimilation or machine learning?"

October 2020

SEMINAR

Joint ECMWF/ESA workshop on ‘Machine Learning for Earth System Observation and Prediction

April 2020

MEDIA

"AI and machine learning at ECMWF"

April 2020

SEMINAR

Seminar series on machine learning

December 2019

SEMINAR

ECMWF Council Lecture by Peter Düben on the future of machine learning in weather forecasting
November 2019

SEMINAR

1st Artificial Intelligence for Copernicus Workshop


  • No labels