PredictWind Expands Weather Model Suite, Debuting AI Forecasting Model (PWAi) in Beta

Major Model Release, Featuring PWAi in Beta, AIFS, and ICON, Sets a New Standard for Global Forecasting Precision and Confidence

AUCKLAND, NEW ZEALAND, November 12, 2025 /EINPresswire.com/ -- PredictWind, the global leader in marine weather forecasting, today announced a major upgrade to its weather model suite, introducing three powerful new global models: PWAi, AIFS, and ICON. The centerpiece of this launch is PWAi, PredictWind’s proprietary AI-powered model, currently in Beta.

Driven by the rapid evolution of AI, this new approach to global weather forecasting combines advanced machine learning while leveraging the proven reliability of physics-based models. The new weather model suite means users will have the most advanced forecast data for a clearer picture and increased confidence in every decision.

“We’re proud to be at the forefront of this AI evolution by launching PWAi in Beta. Early results are strong for accuracy, and future gains will come from combining this technology with the high-resolution observation analysis as it’s released,” said Jon Bilger, PredictWind Founder. “Our goal is to empower users with the most advanced and reliable forecasting technology available, and this release is key to that mission.”

The New Model Suite Explained

PredictWind expands its model suite with three carefully selected global models - PWAi, AIFS, and ICON - enabling users to compare top-performing models for greater clarity and confidence in marine planning. The three major additions are:

- PWAi (PredictWind AI): This is PredictWind's own proprietary global model. It combines the strengths of ECMWF, AIFS, Fengwu, and GraphCast to generate a single, optimized forecast covering the globe. It provides 1-hour time steps for the first three days, giving greater time-step precision than other AI models. PWAi performs strongly in validation studies in the short to medium term over larger-scale areas. It will be a key reference model for establishing consensus across the PredictWind suite.
-AIFS: The ECMWF AIFS model uses advanced AI to deliver timely, accurate global forecasts. It performs especially well offshore, capturing large-scale weather patterns in the medium to longer term, giving mariners a clear, data-driven view of changing conditions.
-ICON: Known for its high-resolution accuracy in Europe, the model produced by the German Weather Service (DWD) is now available worldwide.

These new additions work in parallel with PredictWind's existing lineup, including the high-resolution proprietary PWG/PWE models, the GFS, and the ECMWF, while simultaneously retiring the Spire model, which is being deprecated by Spire.

“No single model tells the whole story. By enabling the comparison of innovative models like PWAi, AIFS, and ICON, we are giving our global community more confidence in every decision, from daily planning to ocean passages.”

The new models are available on the PredictWind App and Website. For details on the models, visit www.predictwind.com/three-new-weather-models. PredictWind Meteorologist, Arnaud Monges, will host a webinar on November 13 to walk users through how these models work and what they mean; register interest here.

ENDS

About PredictWind
PredictWind is the world’s leading provider of marine weather forecasting solutions. Trusted by sailors and mariners worldwide, PredictWind’s technology combines advanced weather data with cutting-edge tools for accurate, reliable, and comprehensive marine weather information.

Editor’s Notes:
Please find supporting imagery at the link here
For more information please contact:

Casey Hodges, Senior Account Director, Roam Generation
casey@roamgeneration.com | +64 11 282 582

Casey Hodges
Roam Generation
casey@roamgeneration.com
Visit us on social media:
Instagram
Facebook
YouTube
X

Legal Disclaimer:

EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.