All news

Google DeepMind Introduces WeatherNext 3 for Hourly Global Weather Forecasts

WeatherNext 3 uses real-time satellite data and weather-station observations to generate hourly forecasts, with up to 5 km resolution for temperature and moisture.

Blue WeatherNext 3 title card for Google DeepMind weather forecasting AI
WeatherNext 3, Google DeepMind’s AI weather forecasting model.

Google DeepMind and Google Research have introduced WeatherNext 3, a global AI weather model that generates new forecasts every hour using live geostationary satellite mosaics and weather-station observations.

Higher detail and more frequent updates

WeatherNext 3 provides temperature and moisture forecasts at resolutions of up to 5 kilometers. Other surface variables are available at 10 kilometers, while atmospheric variables such as wind are generated at 25 kilometers. Google describes the overall picture as roughly five times sharper than WeatherNext 2.

Designed for energy and infrastructure

The model also forecasts 100-meter wind speeds, cloud cover and solar radiation. Those variables can support wind and solar power planning, grid operations, agriculture and logistics.

Google reports up to a 60% improvement in the CRPS metric against IMERG for medium-range precipitation forecasts. This is an improvement against a specific evaluation baseline, not a claim that every forecast is 60% more accurate.

Availability

WeatherNext 3 is integrated across Google Search, Gemini, Google Maps and the Google Maps Platform Weather API. Google also points to access through its cloud products.

Source: Google’s official announcement

Note: Use your national meteorological service for official forecasts, severe-weather warnings and public-safety advisories.

All news

News

© 2026 EzraTech. All rights reserved.

News
EzraTech Consultant