WeatherNext 2 beat NHC forecasts. Its open-source code could power on-chain weather insurance and prediction markets.
Alpha Score of 73 reflects strong overall profile with moderate momentum, moderate value, strong quality, strong sentiment.
On the 2025 Atlantic hurricane season, tests pitted Google DeepMind's latest weather model against the National Hurricane Center's standard forecasting tools. The model – WeatherNext 2 – produced more accurate predictions of both the track and intensity of tropical cyclones, the researchers said.
WeatherNext 2 runs 8 times faster than its predecessor. It generates probabilistic ensemble forecasts out to 15 days with hourly resolution. The researchers noted that the model achieves this performance on lower-resolution data than traditional methods require, and they do not fully understand why.
Earlier members of the WeatherNext family, including GraphCast and GenCast, are open-source. DeepMind has indicated it will release WeatherNext's pretrained weights and code to the public as well.
That open-source decision could affect crypto's weather-linked financial products. Prediction markets let users bet on temperature or hurricane landfalls. Parametric insurance protocols automatically pay out when a defined threshold – 75 mph wind, 2 inches of rain – is crossed. Both rely on timely, accurate data.
An open-source model that outperforms the NHC eliminates the need for licensed data feeds. DeFi developers can integrate WeatherNext 2's output directly, or build derivative products on its weights, without negotiating with Google. The institutional validation from the NHC partnership adds credibility beyond academia.
Adoption is not guaranteed. The chain between model output and smart contract execution still needs oracles that publish predictions on-chain and risk models calibrated to the new forecast resolution. Liquidity has to form around any new derivative product. DeepMind solved the forecasting part. The rest is on the crypto side.
The researchers said the NHC partnership gives the model a stamp of approval that speeds acceptance in regulated industries like reinsurance and commodity trading. The immediate practical outcome may not be a new DeFi product but a shift in how insurers price hurricane risk.
DeepMind plans to continue refining the model, the researchers said.
Drafted by a large language model from the source reporting linked above, then screened by automated publishing checks. It is not read by a journalist before publication. Some articles cite our Alpha Score. Verify prices and figures against the original source. Educational coverage, not personalized advice.