Swiss startup Jua has launched an AI-powered weather forecaster that it says beats leading models from tech giants \u2014 potentially making it the world\u2019s most accurate weather forecasting system.<\/span><\/p>\n Jua claims its model \u2014 dubbed EPT-2 \u2014 is faster and more accurate than both Microsoft\u2019s Aurora and Google DeepMind\u2019s Graphcast. In separate, <\/span>peer-reviewed<\/span><\/a> studies<\/span><\/a>, both of those models were shown to be more accurate than the European Centre for Medium-Range Weather Forecasts (ECMWF)\u2019s ENS forecast, widely regarded as the world leader.<\/span><\/p>\n Jua backs up its bold claims with a new report, published today, that puts EPT-2 head-to-head with top-tier models \u2014 including Aurora and two of ECMWF\u2019s best: ENS and IFS HRES.<\/span><\/p>\n According to the paper, EPT-2 came out on top, delivering the most accurate forecasts across the board. It beat Aurora on key variables like 10-metre wind speed and 2-metre air temperature over a 10-day period, ran forecasts 25% faster, and posted the lowest error scores of all models tested. Jua says it achieved all this while using 75% less computing power than Aurora, the second most efficient system tested.\u00a0<\/span><\/p>\n TNW City Coworking space – Where your best work happ<\/a>ens<\/p>\n A workspace designed for growth, collaboration, and endless networking opportunities in the heart of tech.<\/p>\n<\/div>\n<\/div>\n The research is due to be published on the open-access archive arXiv next week, according to Jua.<\/span><\/p>\n DeepMind\u2019s Graphcast model was not included in the study. Nevertheless, Marvin Gabler, Jua\u2019s CEO and co-founder, is confident that it can beat all of the competition.\u00a0<\/span><\/p>\n \u201cWe respect players like Microsoft Aurora, GraphCast, and Tomorrow.io, but they\u2019re either too slow, too narrow, or still reliant on legacy infrastructure,\u201d said Gabler.\u00a0<\/span><\/p>\n AI-based weather forecasting has been making waves in recent years, driven by demand for more accurate and cheaper ways to predict the Earth\u2019s climate.\u00a0<\/span><\/p>\n Traditional weather models, like those from ECMWF or NOAA, use complex physics equations run on billion-dollar supercomputers. AI models skip the equations, learning patterns from massive datasets, potentially making accurate forecasts thousands of times faster on far cheaper, less energy-intensive machines.<\/span><\/p>\n However, Gabler says Jua takes it a step further than previous AI-based forecasters. <\/span>\u201cWhile others are retrofitting AI onto legacy systems, we\u2019ve built a native physics simulation that understands how Earth\u2019s atmosphere actually behaves,\u201d he said.<\/span>\u00a0<\/span><\/p>\n Jua released its first global AI weather model three years ago. The startup has since raised a total of $27mn in funding from backers including 468 Capital, Future Energy Ventures, and Promus Ventures. <\/span><\/p>\n<\/p><\/div>\n If you liked the article, do not forget to share it with your friends. Follow us on\u00a0Google News<\/a><\/span>\u00a0too, click on the star and choose us from your favorites.<\/span><\/strong><\/p><\/blockquote>\n If you want to read more like this article, you can visit our Technology category.<\/a><\/span><\/strong><\/p>\n<\/blockquote>\n\n
![](\"https:\/\/img-cdn.tnwcdn.com\/image?fit=796%2C417&url=https%3A%2F%2Fcdn0.tnwcdn.com%2Fwp-content%2Fblogs.dir%2F1%2Ffiles%2F2024%2F12%2Fgoogle-deepmind-AI-weather-forecast-gencast.jpg&signature=c4416721ebc9d1389220b9f7122c6901\")
<\/p>\n