Google DeepMind and Google Research have unveiled WeatherNext 3, a new AI weather forecasting system the company says is its most accurate global model to date, and it’s already rolling out across Search, Gemini, Maps, and Google Cloud. Because they lacked fine-grained detail and struggled to fold in real-time data from sources like satellites, while AI has already sped up and sharpened forecasting in recent years, predicting fast-moving, hyper-local weather has remained a weak spot for existing models, largely. Weather shapes an enormous range of everyday and high-stakes decisions, from whether to carry an umbrella to how supply chains, farms, and power grids respond to storms, heatwaves, and droughts.
The resolution gains are substantial: WeatherNext 3 can resolve surface variables like temperature and moisture down to a 5-kilometer grid, other surface variables at 10 kilometers, and atmospheric variables such as wind speed at 25 kilometers — roughly five times sharper overall than WeatherNext 2’s uniform 25-kilometer, 6-hour-increment forecasts. WeatherNext 3 began powering weather features in Google Search, the Gemini app, Google Maps, the Google Maps Platform Weather API, and Google Earth Engine starting the day of the announcement. Google says forecasts a day or more out will be up to 50% more accurate on precipitation specifically, with the largest gains in regions where forecasting has traditionally been weakest.
Google notes that the model doesn’t replace official sources for severe weather warnings or public safety advisories, which it says should still come from local meteorological agencies and national weather services. Independent live evaluations from Brightband currently rank it as the top-performing global weather model available. Developers and researchers can also access the underlying hourly, high-resolution data directly through BigQuery, Earth Engine, or bulk downloads from Google Cloud Storage.