Concluded·following since 6 Aug 2026·Day 3·7 sources·updated 9 Aug 2026

Google DeepMind WeatherNext cyclone forecasting models

The gist

Google DeepMind's WeatherNext adds one day of cyclone lead time, per Nature paper

An extra day of cyclone warning time gives communities more time to prepare for floods and landslides. The open-source release under commercially usable licenses makes the models broadly available.

The full picture

Google DeepMind's WeatherNext AI model gives cyclone forecasters an average of 24 additional hours of lead time compared to existing methods, meaning three-day forecasts are as accurate as prior models' two-day forecasts. The research was published in Nature on August 6, 2026. Google DeepMind described that gain as corresponding to 'roughly a decade's worth of meteorological progress'. During the 2025 hurricane season, WeatherNext predicted with 80% confidence that Hurricane Melissa would make landfall in Jamaica as a Category 5 storm five days before it did. The model was co-trained on nearly 20 terabytes of global atmospheric data and the IBTrACS historical cyclone database covering nearly 5,000 storms. It uses Functional Generative Networks to produce ensemble forecasts, scaling to 1,000 predictions per run, to cover rare events such as rapid intensification. WeatherNext Cyclones operates at a 28x28km resolution, which Google DeepMind said is 100 times coarser than traditional models require. Google DeepMind open-sourced WeatherNext 2, WeatherNext Cyclones, and WeatherNext 2-mini on GitHub under commercially usable licenses. WeatherNext 2-mini runs on a single TPU in a free public Colab notebook.

How it developed

9 August 2026

Google DeepMind published a Nature paper on August 6 showing WeatherNext gives cyclone forecasters 24 additional hours of lead time, a gain it called 'roughly a decade's worth of meteorological progress'.

During the 2025 hurricane season, WeatherNext predicted five days ahead with 80% confidence that Hurricane Melissa would make landfall in Jamaica as a Category 5 storm. Code and weights for all three variants are now on GitHub under commercially usable licenses, with WeatherNext 2-mini runnable in a free Colab notebook.

8 August 2026

Ars Technica reported on WeatherNext, covering the Hurricane Melissa prediction and one-day lead time result

Google DeepMind confirmed that code and weights for WeatherNext 2, WeatherNext Cyclones, and WeatherNext 2-mini are publicly available on GitHub under commercially usable licenses, paired with a Nature paper published August 6 showing the model adds an average of 24 hours of cyclone lead time over existing methods. WeatherNext Cyclones, the variant that operated during the 2025 hurricane season, predicted with 80% confidence that Hurricane Melissa would strike Jamaica as a Category 5 storm five days before landfall. Google DeepMind called that gain 'roughly a decade's worth of meteorological progress'.

First citedDeepMind Blog Google DeepMind Twitter Ars Technica AI deepmind.google+3

6 August 2026

Google DeepMind published WeatherNext Nature paper and open-sourced WeatherNext 2, WeatherNext Cyclones, and WeatherNext 2-mini on GitHub

Sources

Ars Technica AI