Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 1 min ago
Peer-reviewedCC BYGeoscientific model development

AIFS Single 1.1.0: an update to ECMWF's machine-learned weather forecast model AIFS

  • Gabriel Ionut Moldovan1
  • Ewan Pinnington1
  • Ana Prieto Nemesio
  • Simon T. K. Lang1
  • Zied Ben Bouallègue
  • Jesper Sören Dramsch2
  • Mihai Alexe2
  • Mario Santa Cruz1
  • Sara Hahner2
  • Harrison Cook1
  • Helen Theissen1
  • Mariana Clare2
9 more authors
  • Cathal O’Brien2
  • Jan Polster2
  • Linus Magnusson1
  • Gert Mertes1
  • Florian Pinault2
  • Baudouin Raoult1
  • Patricia de Rosnay1
  • RICHARD M. FORBES1
  • Matthew Chantry1
  1. 1European Centre for Medium-Range Weather Forecasts, Shinfield Park, Reading, RG2 9AX, United Kingdom
  2. 2European Centre for Medium-Range Weather Forecasts, Robert-Schuman-Platz 3, 53175 Bonn, Germany
Open at sourceDOICite

Abstract

We present an update to ECMWF's machine-learned weather forecasting model AIFS Single with several key improvements. The model now incorporates physical consistency constraints through bounding layers, an updated training schedule, and an expanded set of variables. The physical constraints substantially improve precipitation forecasts and the new variables show a high level of skill. Upper-air headline scores also show improvement over the previous AIFS version. The AIFS has been fully operational at ECMWF since the 25th of February 2025.

Related

  1. Forecast Skill Is Not Decision Skill: Evidence From Weather‐Dependent Decision Tasks
    Kornelius Raeth, Nicole Ludwig
    Meteorological ApplicationsPeer-reviewedCC BYFull textAtmospheric science & meteorologyDOAJ
  2. Forecast lead time and the timing of crew deployment for weather-driven power outages using AI weather models
    Mohammad Kian Golkar, Luciano Alves de Oliveira, Mohamad Alipour
    arXivPreprintCC BY 4.0Full textAtmospheric science & meteorologyarXiv
  3. Can AI Forecast the Extremes? Evaluating Probabilistic Weather Models on Tropical Cyclones
    Nathan Chalumeau, Xavier Renard, Remi Meynadier +3
    arXivPreprintCC BY 4.0Full textAtmospheric science & meteorologyarXiv
  4. Forecast‐Error Diagnostics in Neural Weather Models
    Uroš Perkan, Gregor Skok, Žiga Zaplotnik
    Quarterly Journal of the Royal Meteorological SocietyPeer-reviewedCC BY5 citedAtmospheric science & meteorologyOpenAlex
  5. Every Fixed Metric Has a Blind Spot: A Learned Atmospheric Critic for Scoring Forecast Realism
    Younes Elberkennou, Dmitri Demler, Thierry Meier +3
    arXivPreprintCC BY 4.0Full textAtmospheric science & meteorologyarXiv

Cite

BibTeXRIS
BibTeX
@article{moldovan2026aifs,
  title = {{AIFS Single 1.1.0: an update to ECMWF's machine-learned weather forecast model AIFS}},
  author = {Gabriel Ionut Moldovan and Ewan Pinnington and Ana Prieto Nemesio and Simon T. K. Lang and Zied Ben Bouallègue and Jesper Sören Dramsch and Mihai Alexe and Mario Santa Cruz and Sara Hahner and Harrison Cook and Helen Theissen and Mariana Clare and Cathal O’Brien and Jan Polster and Linus Magnusson and Gert Mertes and Florian Pinault and Baudouin Raoult and Patricia de Rosnay and RICHARD M. FORBES and Matthew Chantry},
  journal = {Geoscientific model development},
  year = {2026},
  month = jun,
  doi = {10.5194/gmd-19-4703-2026},
  url = {https://doi.org/10.5194/gmd-19-4703-2026},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - AIFS Single 1.1.0: an update to ECMWF's machine-learned weather forecast model AIFS
AU  - Gabriel Ionut Moldovan
AU  - Ewan Pinnington
AU  - Ana Prieto Nemesio
AU  - Simon T. K. Lang
AU  - Zied Ben Bouallègue
AU  - Jesper Sören Dramsch
AU  - Mihai Alexe
AU  - Mario Santa Cruz
AU  - Sara Hahner
AU  - Harrison Cook
AU  - Helen Theissen
AU  - Mariana Clare
AU  - Cathal O’Brien
AU  - Jan Polster
AU  - Linus Magnusson
AU  - Gert Mertes
AU  - Florian Pinault
AU  - Baudouin Raoult
AU  - Patricia de Rosnay
AU  - RICHARD M. FORBES
AU  - Matthew Chantry
PY  - 2026
DA  - 2026/06/01
JO  - Geoscientific model development
DO  - 10.5194/gmd-19-4703-2026
UR  - https://doi.org/10.5194/gmd-19-4703-2026
AB  - We present an update to ECMWF's machine-learned weather forecasting model AIFS Single with several key improvements. The model now incorporates physical consistency constraints through bounding layers, an updated training schedule, and an expanded set of variables. The physical constraints substantially improve precipitation forecasts and the new variables show a high level of skill. Upper-air headline scores also show improvement over the previous AIFS version. The AIFS has been fully operational at ECMWF since the 25th of February 2025.
SN  - 1991-959X
SN  - 1991-9603
ER  -