Google DeepMind Launches WeatherNext 2, a Breakthrough AI Weather Forecasting Model
Google DeepMind has unveiled WeatherNext 2, an advanced artificial intelligence model designed to revolutionize weather prediction and extreme weather forecasting. The new model represents a significant leap forward in AI-driven meteorology, delivering faster and more accurate forecasts than previous iterations.
Key Capabilities and Performance Improvements
WeatherNext 2 is eight times faster than its predecessor, enabling the analysis of significantly more weather scenarios per forecast. This enhanced speed translates into better predictions for low-probability but catastrophic weather events—critical information for emergency preparedness and risk management.
The model can forecast crucial weather variables with high accuracy, including wind speed and direction, precipitation, and atmospheric pressure. It generates six-hour forecasts four times daily, providing timely hour-by-hour data that supports improved decision-making across various sectors.
Cyclone Prediction and 15-Day Forecasts
One of WeatherNext 2’s standout features is its ability to predict the likely path, intensity, structure, and size of cyclones—including where they will form—up to 15 days in advance. This extended prediction window offers valuable lead time for communities and emergency responders to prepare for potential impacts.
Widespread Integration and Availability
The WeatherNext 2 model family is now integrated into Google’s core forecasting system, powering weather features across Search, Gemini, Pixel Weather, and Google Maps, as well as the platform’s Weather API. This integration ensures that millions of users have access to more accurate weather information.
For businesses and enterprises, WeatherNext 2 models are available through Google Cloud Vertex AI, allowing organizations to customize forecasts for their specific needs. Researchers and scientists can also access the model for advanced research and geospatial analysis through BigQuery and Google Earth Engine.
Advancing Weather Science
By making WeatherNext 2 available to users, researchers, and enterprises, Google DeepMind is supporting better decision-making in an increasingly unpredictable climate. The model exemplifies how artificial intelligence can enhance traditional meteorological science, offering faster processing and more comprehensive scenario analysis than conventional forecasting systems.