- Nvidia Earth-2 speeds up weather forecasting and significantly reduces computational costs
- Earth-2 includes CorrDiff, FourCastNet3, Medium Range, Nowcasting, Global Data Assimilation, and the PhysicsNeMo framework.
- Energy companies rely on Earth-2 to improve grid reliability and PV predictions
Nvidia has unveiled its new Earth-2 family of open AI models, which it claims could transform weather forecasting and climate prediction as we know it.
The Nvidia Earth-2 family includes CorrDiff, FourCastNet3, Medium Range, Nowcasting, Global Data Assimilation, and the PhysicsNeMo framework for training and tuning AI physics models.
These models integrate high-resolution data from satellites, radars and weather stations to provide continuous estimates of atmospheric conditions.
High resolution modeling for fast forecasts
Earth-2 uses generative AI to accelerate every stage of forecasting, from processing observational data to generating global and localized storm predictions.
CorrDiff uses a generative AI architecture to reduce broad continental predictions to high-resolution regional forecasts, producing results up to 500 times faster than traditional methods.
FourCastNet3 delivers accurate wind, temperature and humidity forecasts, outperforming conventional ensemble models and providing up to 60 times faster predictions.
The system also integrates models from the European Center for Medium-Range Weather Forecasts, Microsoft and Google, allowing users to combine multiple approaches within a single framework.
Nvidia’s PhysicsNeMo enables AI physics models to be trained and tuned at scale, offering flexibility for both operational forecasting and scientific research.
Earth-2’s global data assimilation produces initial atmospheric conditions in seconds on GPUs instead of hours on supercomputers, allowing for faster integration into downstream models.
Organizations in the research, energy, and government sectors are already using these AI tools to improve forecast accuracy and reduce computational costs.
The Israel Meteorological Service already uses CorrDiff and plans to implement Nowcasting for high-resolution predictions up to eight times a day.
Energy companies such as TotalEnergies, Eni and GCL are testing Earth-2 to improve grid operations, short-term risk awareness and PV forecasting.
Brightband and Taiwan meteorologists use Earth-2 CorrDiff and Medium Range to deliver accurate global and local forecasts, and The Weather Company is now evaluating Nowcasting for ultra-short-term local storm predictions.
These AI tools reduce computational demand, with some models reporting a 90% reduction in computing time compared to classical methods on CPU clusters.
The open source availability of Earth-2 on platforms such as Hugging Face and GitHub allows researchers, companies and startups to adjust forecasts for local conditions.
By combining multiple AI models and tools, organizations can generate probabilistic, actionable insights that inform decisions in agriculture, energy, disaster response, and insurance risk assessment.
“Philosophically and scientifically, it’s a return to simplicity… We’re moving away from hand-crafted niche AI architectures and toward the future of simple, scalable transformative architectures,” said Mike Pritchard, director of climate simulation at Nvidia.
“This provides the building blocks used by everyone in the ecosystem (national weather services, financial services companies, energy companies) and anyone who wants to build and refine weather forecast models.”
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds. Be sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp also.
