Google Puts WeatherNext 3’s Hourly AI Forecasts Into Search, Maps and Gemini
The launch moves an observation-led weather model from Google’s research stack into consumer products and cloud data tools. Its performance claims are substantial, while Google directs users seeking official severe-weather warnings and public-safety advisories to meteorological agencies.
Listen to this story
The audio brief
Story brief
3 key pointsWeatherNext 3 moves Google’s latest weather model from research release into a broad product rollout, with hourly global updates and variable-specific resolution down to 5 kilometers. Compared with WeatherNext 2’s six-hour, 25-kilometer forecasts, Google reports sharper output and improved precipitation scores, though these are company evaluations rather than safety guarantees. Wind, cloud and solar-radiation data...
- 01
Temperature and moisture forecasts use a 5-kilometer grid; other surface variables use 10 kilometers, and atmospheric measures 25 kilometers.
- 02
Google reports early-lead precipitation CRPS improvements of up to 60% versus IMERG, 30% versus MRMS, and 10% versus rain gauges.
- 03
At least one day ahead, precipitation forecasts are reportedly up to 50% more accurate, especially in historically weaker regions.
Google Research and Google DeepMind have introduced WeatherNext 3, an AI weather model that generates new global forecasts every hour from live geostationary satellite data and historical analysis. Google says it will begin powering weather experiences worldwide across Search, Gemini, Maps, Maps Platform’s Weather API and Earth Engine. The practical wager is that faster refreshes and finer local detail can make a general-purpose product layer more useful for rain, wind and energy planning.
An hourly forecast built from observations
The model also trains directly on sparse weather-station observations to account for local terrain. It produces temperature and moisture forecasts on a 5-kilometer grid, other surface variables at 10 kilometers, and atmospheric measures such as wind speed at 25 kilometers. That variable-by-variable design means the headline 5-kilometer figure does not apply to every output.
Precipitation is a key test
Google reports improvements in Continuous Ranked Probability Score, a measure for probabilistic forecasts, of up to 60% against IMERG, 30% against MRMS and 10% against rain-gauge measurements at early lead times. It also says forecasts made at least a day ahead are up to 50% more accurate for precipitation, with the largest gains in places where forecasts have historically been less reliable. Those are company-reported evaluation results, not a public-safety guarantee.
One model, several routes to use it
It also adds variables designed for renewable-energy planning: 100-meter wind speeds, roughly turbine height, plus cloud cover and solar radiation. Google says those inputs can help estimate wind and solar output and match expected generation with demand. The same detail could be useful across many operational settings, but the release does not establish results for a specific grid operator, business or region.
Official guidance remains with weather agencies
Editorial analysis
Our Read
WeatherNext 3 is notable less as a standalone model release than as a distribution move. Google is placing the same forecast layer in consumer surfaces, developer infrastructure, geospatial tools and renewable-energy planning inputs. That could make forecast quality an increasingly integrated Google service rather than a specialist product. The key evidence to watch next is whether Google’s reported precipitation gains hold in independent live evaluations and whether the hourly, higher-resolution data improves decisions in the regions Google identifies as historically less reliable for forecasting.
Sources
- blog.googleIntroducing WeatherNext 3, our most advanced and accurate global weather AI model
