An extra day of warning: DeepMind opens its cyclone model, and the Hurricane Center is already using it
Google DeepMind published research in Nature on August 6 showing that its WeatherNext model predicts a tropical cyclone's track, intensity and wind structure more accurately than existing methods: its three-day forecasts are about as accurate as two-day forecasts used to be, a jump the team compares to a decade of ordinary progress. The model was trained on roughly 20 terabytes of weather data and the IBTrACS archive of past storms, and produces 1,000 possible futures per forecast; a 15-day run takes under a minute on a single TPU chip. The US National Hurricane Center used it operationally through the 2025 season, including on Hurricane Melissa's rapid intensification before it hit Jamaica. DeepMind has now released code and model weights for WeatherNext Cyclones, WeatherNext 2 and a small WeatherNext 2-mini on GitHub under the Apache 2.0 licence; the mini version runs in a free Colab notebook.
Related stories
Machine learning read the shape of sick brain cells and picked out nine already-approved drugs that calmed them down2026-08-21Given four hours and a GPU to improve the way AI is trained, the best agent scored 0.25 out of 1 — and most never tried
2026-08-21The agent invented a second person to vouch for its code. A 24-year-old in Texas refused to believe either of them.
2026-08-21Pew put half a million web pages through a detector: a third of everything published since ChatGPT carries its marks
2026-08-21The FDA has cleared 1,357 AI medical devices. Three of them have been tested on whether patients live longer or better.
2026-08-20