Cordulus boosts local weather forecast accuracy by 63% with AI model trained on Gefion

Danish weather tech company Cordulus leverages sovereign AI infrastructure to deliver hyper-local, high-precision weather forecasts.

Cordulus has developed a new AI weather model capable of predicting rain with unprecedented precision, achieving a 63% increase in accuracy compared to traditional forecasts. This leap forward allows industries to make better, faster decisions in sectors such as agriculture, construction, and energy.

“We all rely on weather forecasts every day, but they can also be a source of frustration,” says John Smedegaard, CEO of Cordulus. “Our goal is to provide a new level of confidence in the weather. We believe this is essential not only for countless business decisions but also for adapting to a changing climate. Each day, we collect and process enormous amounts of weather data that we use to power our AI forecasts. Gefion allows our AI to learn from these real-world events and become better at predicting the future.”

Unlike generic forecasts, AI-powered weather models can deliver ultra-precise, street-level predictions, which are increasingly critical as extreme weather events become more frequent and severe.

“It’s inspiring to see how Gefion’s capacity to process vast amounts of data and train advanced models at scale, combined with Cordulus’ unique weather data and innovative AI technology, can produce forecasts that reach new levels of accuracy and reliability. We now have several clients using Gefion to model weather data with impressive results” says Ali Syed SVP of DCAI.

About Cordulus
Cordulus is a Danish weather technology company that provides hyper-local weather insights using innovative AI and advanced weather technology.
Cordulus has more than 10 years’ experience in providing trusted weather information, and today has clients in more than 23 countries.