LA JOLLA, Calif., Sept. 23, 2025 – Technosylva, the global leader in wildfire science and technology, today announced the launch of the world’s largest dedicated supercomputers for wildfire modeling. The system simulates more than one billion fire scenarios each day, applies artificial intelligence to decades of fire weather data, and can more accurately identify which communities and electric lines may be threatened up to five days in advance. Results are delivered in seconds, giving utilities and fire agencies critical lead time to protect lives and infrastructure.

As wildfires intensify across the country, the scale of the threat now demands computing power to match. More than 70,000 U.S. communities — representing 46 million homes — are at risk in the wildland–urban interface. Over the past decade, wildfires have destroyed more than 35,000 structures, according to the U.S. Forest Service. Without high-performance computing and AI, agencies are left with slower, less precise models using insufficient public source data that cannot capture the complexity of modern wildfire behavior, increasing the risk of failed ignition prevention, unnecessary power shutoffs, or inadequate emergency response. For extreme weather applications, it also creates the kind of on-time intelligence required for hurricanes, floods and other weather events to restore power quickly and position emergency resources and mitigations. 

“Accurate and timely wildfire forecasting requires supercomputing at scale,” said Joaquin Ramirez, CTO and Founder of Technosylva. “This platform fuses high-performance computing with advanced AI, running ensemble forecasts against decades of meteorological and fire-behavior data. The result is faster, more precise predictions that set a new benchmark for operational wildfire readiness.”

Developed with PSSC Labs, the 11,500-core system is purpose-built for fire behavior forecasting. It incorporates comprehensive wildfire data beyond what standard, publicly available weather and fuel models provide, including live and dead fuel moisture, vegetation conditions, terrain-driven winds, and three decades of historical fire weather patterns. Forecasts are generated at two-kilometer resolution and draw on terabytes of data - thousands of critical fire variable data feeds – processed every few hours. The system runs many wildfire models in parallel and in conjunction, including refined, proprietary models for fire spread, crown spotting, extreme weather, fuel moisture and urban conflagration that Technosylva pioneered and that have now become critical to understanding and forecasting today’s complex and dangerous wildfires. That scale allows utilities and agencies to make precise power shutoff decisions, accelerate evacuation alerts, and strengthen resilience across the country.

“This platform translates faster computing power and groundbreaking science into life-saving and actionable intelligence that any utility or fire agency can understand and use,” said Brent Shaw, Senior Numerical Weather Prediction Architect at Technosylva. “We can process massive datasets in hours and deliver forecasts in seconds. It is like upgrading from an X-ray to a real-time MRI of wildfire behavior across the entire country. The same system also makes it possible to train and deploy AI models that capture wildfire dynamics in real-time.”

The new supercomputer accelerates Technosylva’s research and development, creating a foundation for faster integration of artificial intelligence and machine learning into operational wildfire products.

“AI depends on both scale and speed,” said Ramirez. “This infrastructure gives us the ability to train and validate models with unprecedented amounts of data, and then turn those insights into practical tools for utilities and first responders.”

PSSC Labs has supported Technosylva’s growth as it has continued to expand its coverage areas, data inputs, accuracy and results granularity.

“We have spent 25 years perfecting high-performance computing solutions for real-world applications that save lives,” said Alex Lesser, Vice President of Sales and Marketing at PSSC Labs. “This dedicated platform ensures utilities and fire agencies have access to the most advanced wildfire modeling available, without waiting for resources to free up.”

The expanded HPC system is now operational and available for utility and agency partners nationwide.