- WeatherNext 2 uses a Functional Generative Network to create coherent and realistic weather scenarios.
- Generate hundreds of forecasts in less than a minute with a single TPU, with hourly resolution.
- It improves accuracy to 99,9% of variables between 0 and 15 days and is already integrated into Search. Gemini: and Pixel.
- Data available in Earth Engine and BigQuery, and early access in Vertex AI for custom models.
If you're one of those people who checks the weather forecast before leaving home, you'll know that weather apps are almost as essential as your phone itself. Among the most popular are BBC Weather, Met Office, Apple Weather, The Weather Channel, and AccuWeather, all very reliable for planning your day. However, what's coming with Google 's new model promises to raise the bar: WeatherNext 2 aims to speed up, refine, and make the forecasts we consult constantly much more useful.
The result of a joint effort between Google DeepMind and Google Research, this AI system aims to generate forecasts eight times faster than traditional approaches, with a temporal resolution of up to one hour and the ability to simulate hundreds of scenarios from a single starting point. This leap is already noticeable in products you use daily, as forecasts in Search, Google Gemini , Pixel Weather, and the Google Maps Platform API have been updated to take advantage of the new technology, with more features on the way for Google Maps.
What is WeatherNext 2 and why is everyone talking about it?
WeatherNext 2 is the second iteration of Google's AI-powered weather forecasting system. Unlike a classic deterministic model, this approach focuses on generating multiple plausible scenarios that respect the physics of the atmospheric system. Starting with a single initial weather state, it can produce a range of outcomes in under a minute, helping to anticipate both the most likely weather patterns and low-probability extremes —key to risk management.
The change is also significant operationally: instead of relying on calculations that run for hours on supercomputers with numerical models, WeatherNext 2 runs its inferences on a TPU (Tensor Processing Unit) and obtains hundreds of forecasts in record time. This efficiency isn't just cosmetic: it means faster decisions when urgent action is needed, for example, in the face of severe storms or winds that affect critical infrastructure.
How it works: the Functional Generative Network behind the model
The technical heart of the system is called the Functional Generative Network (FGN) . This architecture injects controlled "noise" directly into the model's functional space, helping simulations maintain physical consistency and interdependence between variables. In practice, instead of fixing a single trajectory, the model generates several possible paths for the atmospheric state that fit together and are consistent with the system's own laws.
An intuitive way to understand this is to think of an orchestra rehearsing the same piece in sections: each group works on its part with slight variations, but all the sections fit together harmoniously when combined. Similarly, FGN allows different networks to learn aspects of weather separately, introduce variability, and still build joint forecasts that are physically consistent.
This approach overcomes the limitations of "marginal" forecasts, which describe a single variable (such as the exact temperature at a specific point and time). The FGN facilitates the transition to integrated predictions where temperature, wind, humidity, and other variables evolve in an interrelated way, which is crucial for capturing complex phenomena and incorporating low-probability extreme events into the range of scenarios.
Another key aspect is that the model is trained using what researchers describe as "loose data" from the climate : measurements of variables such as wind, temperature, and humidity at specific locations. From this single data point, the FGN is able to infer and anticipate the dynamics of larger-scale weather systems , identify areas at risk of heat waves, and estimate how much energy a wind farm could generate under future conditions.
Speed, resolution, and prediction horizons
In terms of metrics, the leap from the previous version is significant: according to Google, WeatherNext 2 outperforms 99,9% of the analyzed variables (temperature, wind, humidity, and more) across all forecast horizons from 0 to 15 days . Furthermore, it offers a temporal resolution of up to one hour, enabling more precise forecasts where extra detail makes a difference, such as in power grid management or logistics operations.
Computational performance is also a game-changer. Where conventional simulations require several hours on supercomputers , WeatherNext 2 calculates hundreds of scenarios in less than a minute using a TPU. And there's an interesting operational nuance: the system processes the most recent atmospheric conditions and, from there, generates four six-hour forecasts per day, incorporating variations thanks to the FGN to cover the range of possibilities in greater detail.
- Covered horizons: from nowcasting and short term up to 15 days, with clear improvements in accuracy.
- Temporary resolution: up to one hour, useful for immediate decisions in sensitive sectors.
- Scenario capacity: hundreds in less than a minute from a single initial state.
- Efficient computing: execution on a TPU, without the need for massive supercomputing.
Integration into Google products and availability to the community
WeatherNext 2 technology is already driving improvements to Search, Gemini, Pixel Weather , and the Google Maps Platform API. Google has also announced that in the coming weeks we will see visible improvements to Google Maps related to weather information, making it easier to check conditions and forecasts at a glance for different scenarios.
Regarding data access, Google states that WeatherNext 2 forecasts are available in Earth Engine and BigQuery . Furthermore, they have launched an early access program on Vertex AI (Google Cloud) that allows companies and technical teams to run custom inferences and fine-tune the model to their needs—a direct path to adoption in real-world environments.
Looking ahead, the Google DeepMind and Google Research teams say they are exploring new capabilities to integrate more data sources, expand access, and continue transforming cutting-edge research into high-impact applications . Their stated commitment is to make these new tools available to the global community to accelerate scientific progress and improve decision-making.
From the labs to everyday life: what changes for the user
The practical difference lies in the fact that, from a single input, WeatherNext 2 can simulate hundreds of possible weather scenarios and show you the most likely ones, as well as the worst-case scenarios to watch out for. This is especially useful for planning trips, flights, commutes, supply chains , or the operations of delivery companies, where small weather variations can lead to costs or delays.
If you already use a Pixel phone, you know its weather app is known for its clarity. With WeatherNext 2, forecasts in Pixel Weather and Search are now more accurate and faster, so you can see key information at a glance, including the range of possibilities instead of a single optimistic figure that changes at the last minute.
Google suggests that this advancement reduces reliance on radar and satellites as they have been used until now, since the model learns patterns from large volumes of historical and recent data. In any case, the beauty of the new system is not to replace everything, but to create a more useful forecast by integrating different signals and quickly addressing uncertainty.
Use cases: from energy to emergency management
- Emergency services: Evaluate in advance scenarios of severe storms, floods or strong winds to activate protocols in time.
- Renewable energy: to estimate future production in wind or solar farms, adjusting supply to demand with greater precision.
- Transport and logistics: Optimize routes and operating windows in the face of rain, snow or gusts of wind, minimizing delays and costs.
- Public sector and insurance: analyze risks in worst case scenarios low probability and high impact to size response and coverage.
- Research and education: explore the dynamics of extreme events and compare joint scenarios to better understand the atmosphere.
What does it offer compared to traditional methods?
Numerical weather models solve physical equations of the atmosphere using high-resolution grids and, by their very nature, require significant computing power . WeatherNext 2's AI doesn't ignore physics: it learns from historical data and the current state to generate plausible scenarios consistent with those laws, but it does so at a lower computational cost and in much shorter times.
Its key advantage is its ability to probabilistically cover the full range of possibilities, including extreme cases that are relevant for planning. Where previously a large set of costly simulations was needed to approximate this uncertainty, FGN introduces "intelligent" variability that is physically consistent, allowing hundreds of trajectories to be explored almost in real time.
Limits, responsibilities and good practice
It's important to remember a basic principle: the atmosphere is a chaotic system . Even with AI, 100% infallible predictions will not be achieved. The danger of a highly accessible and seemingly accurate tool is that the average user will assume certainty where there are only probabilities. Here, it's essential that the interfaces communicate uncertainty and that a minimum level of meteorological literacy is maintained.
Some voices are raising concerns about whether it's a good idea to " put WeatherNext 2 in just anyone's hands " without expert review, just as a cancer-detecting algorithm needs a radiologist to validate the results. The tool is incredibly powerful, but its nuanced interpretation, especially in safety and self-protection situations , still requires professional judgment.
Recent experience with severe weather events demonstrates this. In episodes like the DANA storm in Valencia , delayed decision-making can exacerbate the impact. A system that quickly provides worst-case scenarios and delivers clear warnings buys valuable minutes to activate contingency plans and mitigate risks. This is where improvements in speed and resolution are most transformative.
The key will be combining the best of both worlds: the power of AI to produce rich and rapid scenarios, and the validation of specialists to refine messages and action guidelines. If, in addition, honest communication about ranges and probabilities is maintained, the result will be a better-prepared citizenry without a false sense of certainty.
Access for companies, scientists, and developers
For those who need to work with data at scale, Google has made WeatherNext 2 forecasts available in Earth Engine and BigQuery , facilitating queries, analysis, and cross-referencing with other sources. And for those looking to adapt the model to their specific needs, there's an early access program on Vertex AI with support for custom model inference and deployments in production environments.
The roadmap involves integrating new data sources , expanding access, and continuing to translate research advances into high-impact, real-world applications. With this impetus, agencies and companies can accelerate their transition to data-driven decision-making, where weather , far from being an unpredictable factor, becomes a measurable and manageable risk.
WeatherNext 2 represents a significant leap forward in weather forecasting: eight times faster than traditional approaches, capable of outperforming its predecessor by 99,9% of variables over 0- to 15-day forecast horizons, and ready to generate hundreds of scenarios in under a minute from a single initial state. Integration with Search, Gemini, Pixel Weather, and the Maps API is already underway, while data in Earth Engine and BigQuery, along with early access to Vertex AI, opens the door to professional and scientific applications that, when properly guided, can make a real difference in how we prepare for potential weather events.
Passionate writer about the world of bytes and technology in general. I love sharing my knowledge through writing, and that's what I'll do on this blog, show you all the most interesting things about gadgets, software, hardware, tech trends, and more. My goal is to help you navigate the digital world in a simple and entertaining way.