WeatherNext: AI model achieves breakthrough in forecasting cyclones
Our model uses Functional Generative Networks (FGNs) to efficiently produce ensembles of different predictions, which captures the inherent uncertainty of the weather.
Key points
- We can now generate a single 15-day forecast in less than a minute on a TPU, empowering forecasters to quickly evaluate the probability distribution of potentially devastating tail-risks.
- However, WeatherNext Cyclones only needs data with a resolution of 28x28km, 100x coarser than traditional models.
- Opening up WeatherNext to the research community
- We are also releasing two sets of similar models: WeatherNext Cyclones, which ran during the hurricane season (results can be seen in the paper); and WeatherNext 2, a later update that we operationalized in October.
Sources (1)
- [1]WeatherNext: AI model achieves breakthrough in forecasting cyclonesGoogle DeepMind Blog · Aug 6, 03:06 PM
Our model uses Functional Generative Networks (FGNs) to efficiently produce ensembles of different predictions, which captures the inherent uncertainty of the weather.
We can now generate a single 15-day forecast in less than a minute on a TPU, empowering forecasters to quickly evaluate the probability distribution of potentially devastating tail-risks.
Extractive summary: sentences quoted from the sources.
Before this
- Jul 30, 2026Gemini Robotics ER 2: powering robotics with video understanding, task orchestration, and multi-robot collaboration
- Jul 17, 2026Introducing Gemini 3.5 Flash Cyber
- Jun 30, 2026Start building with Nano Banana 2 Lite and Gemini Omni Flash
- Jun 10, 2026DiffusionGemma: 4x faster text generation
- Jun 9, 2026Introducing Gemma 4 12B: a unified, encoder-free multimodal model
- May 6, 2026AlphaEvolve: How our Gemini-powered coding agent is scaling impact across fields