Google WeatherNext 3 Sets a New Benchmark for AI Weather Forecasting

Google DeepMind and Google Research have introduced WeatherNext 3, an advanced AI-powered global weather forecasting model engineered to deliver localized, real-time climate intelligence at unprecedented resolution.

By directly ingesting live geostationary satellite mosaics and real-world ground station data, WeatherNext 3 eliminates traditional reliance on slow, compute-heavy numerical approximations. The result is an operational system that produces fresh forecasts every hour of the day at up to 5-kilometer spatial resolution—roughly five times sharper than its predecessor.

Key Breakthroughs & Capabilities:
● Continuous Hourly Forecasting: Generates new predictions every hour, providing vital lead time as storms, fronts, and extreme weather develop rapidly.
● Ultra-High Resolution: Resolves surface temperature and moisture down to 5 km (0.05°) and surface winds to 10 km (0.1°), delivering hyper-local precision even in traditionally underserved areas.
● Superior Precipitation Accuracy: Cuts error metrics by up to 50% on key precipitation benchmarks compared to traditional numerical weather prediction baselines.
● Clean Energy Optimization: Outputs specialized parameters—including 100-meter turbine-height wind speeds, multi-layer cloud cover, and solar irradiance—to assist grid operators and clean-energy producers.
WeatherNext 3 has already rolled out powering weather experiences within Google Search, Gemini app, Google Maps, Google Maps Platform Weather API, and Google Earth Engine starting today. The update dramatically improves longer term forecasts. When planning a day or more ahead, people will see up to 50% more accurate precipitation forecasts — with the greatest improvements in regions where forecasts have historically been less reliable.