cd /news/artificial-intelligence/google-deepminds-weathernext-2-outpe… · home topics artificial-intelligence article
[ARTICLE · art-120402] src=cryptobriefing.com ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

Google DeepMind’s WeatherNext 2 outperforms previous AI weather models on 99.9% of variables

Google DeepMind and Google Research released WeatherNext 2 and WeatherNext Cyclones (WN-C), AI weather forecasting models that outperform previous models on 99.9% of forecast variables and deliver forecasts up to 8 times faster than predecessors. A peer-reviewed study in Nature on August 6, 2026, found WN-C provides about one additional day of lead time for cyclone track, intensity, and size predictions compared to conventional systems, with the models open-sourced on GitHub under Apache 2.0 and integrated into Google Cloud via BigQuery and Earth Engine.

read2 min views1 publishedSep 3, 2026
Google DeepMind’s WeatherNext 2 outperforms previous AI weather models on 99.9% of variables
Image: Cryptobriefing (auto-discovered)

Photo: Tima Miroshnichenko / Pexels

The new models predict cyclone tracks a full day earlier than conventional systems and have been open-sourced on GitHub

Google DeepMind and Google Research have released their latest generation of AI weather forecasting models, and the performance gap between machine learning and traditional supercomputer-driven meteorology just got a lot wider.

The updated WeatherNext family, including WeatherNext 2 and a specialized cyclone-tracking variant called WeatherNext Cyclones (WN-C), delivers forecasts up to 8 times faster than its predecessors while outperforming conventional physics-based systems on nearly every measurable dimension. WeatherNext 2 beats previous models on 99.9% of forecast variables.

What the models actually do #

WeatherNext operates at 0.25-degree spatial resolution, which translates to roughly 30 kilometers at the equator. Forecasts initialize every six hours, at 00, 06, 12, and 18 UTC, and ensemble versions extend predictions out to 15 days.

The cyclone-specific model, WN-C, is where things get especially interesting. A peer-reviewed paper published in Nature on August 6, 2026, documents WN-C’s performance during the 2025 Atlantic hurricane season. The headline finding: the AI model provides approximately one additional day of lead time over conventional forecasting systems for cyclone track, intensity, and size predictions. The model proved particularly effective at predicting Hurricane Melissa, one of the more notable storms of the 2025 season.

The models are trained on ERA5 and HRES data from the European Centre for Medium-Range Weather Forecasts. The underlying architecture relies on Functional Generative Networks (FGN) to produce ensemble forecasts, a technique that generates diverse scenario projections while consuming far less computing power than the supercomputers traditionally tasked with weather modeling.

Open source and cloud integration #

The WeatherNext models, including pretrained weights, have been released as open source on GitHub under an Apache 2.0 license. There’s also a mini variant of WeatherNext that can run on a single TPU in Google Colab’s free notebooks.

Google has also integrated the models into its cloud infrastructure, making them accessible through BigQuery and Earth Engine. The phased integration into Google Cloud and consumer-facing products has been underway since 2025.

Why industries beyond tech should pay attention #

Weather forecasting accuracy has direct financial consequences across a surprising number of sectors. Energy markets are the most obvious beneficiary. Grid operators managing renewable energy sources like wind and solar need accurate multi-day forecasts to balance supply and demand. Agriculture is another sector where improved lead times matter enormously. Insurance and reinsurance companies, which price catastrophe risk partly based on forecast model outputs, may need to recalibrate their approaches given a model that can predict cyclone intensity a full day earlier than conventional tools.

The availability of these models through cloud infrastructure and open-source channels also lowers the barrier to entry for startups and smaller enterprises that want to build weather-dependent services.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our

Editorial Policy.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @google deepmind 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/google-deepminds-wea…] indexed:0 read:2min 2026-09-03 ·