- MPAS adaptation for GPUs will anchor the collaboration’s initial work
- TempoQuest CEO Gene Pache calls speed the toughest part of operational forecasting
- TempoQuest’s AceCAST platform contributed to MITRE’s Weather 1K dataset
MITRE and TempoQuest are expanding their partnership to apply weather prediction powered by graphics processing units, or GPUs, to federal decision-making in national security, wildfire response, infrastructure protection and other mission areas.
MITRE said Wednesday the expanded effort will build on Weather 1K, a roughly 7-petabyte AI training dataset that MITRE developed with NVIDIA’s Earth-2 team using tools that included TempoQuest’s AceCAST.
What Will the Expanded Collaboration Explore?
The partners will begin by adapting the Model for Prediction Across Scales, or MPAS, to run on GPUs. MPAS is a weather modeling framework that supports high-resolution forecasting across geographic scales, and the National Oceanic and Atmospheric Administration is incorporating it into its next-generation forecasting systems.
MPAS’ heavy computing demands can slow its simulations, and running it on GPUs could give federal agencies faster access to weather data for time-sensitive decisions.
The collaboration will also look at GPU-accelerated modeling for aviation and transportation operations, critical infrastructure risk management, defense mission planning and readiness, and wildfire and disaster preparedness and response.
MITRE will assess how the capabilities align with federal operational and security requirements, drawing on its systems engineering and mission integration experience. Both organizations intend to engage federal agencies, research institutions and industry partners to test and advance forecasting technologies for operational use.
What Did MITRE & TempoQuest Officials Say About the Collaboration?
Beth Meinert, senior vice president and general manager of MITRE Public Sector, said faster and more accurate weather data can shape how federal, state and local governments prepare for and respond to weather events.
“By combining TempoQuest’s advanced weather modeling capabilities with MITRE’s systems engineering and mission expertise, we help federal agencies and state and local governments translate promising technologies into practical capabilities that strengthen communities and save lives,” she added.
TempoQuest CEO Gene Pache noted that the partnership will help the company focus its model acceleration efforts on federal priorities.
“The hardest part of operational forecasting isn’t the physics – it’s getting the answer fast enough to act on. MITRE understands federal missions in a way no commercial partner can, and working alongside their teams will help us direct our acceleration work at the problems that matter most,” Pache said.
What Other Weather & Advanced Computing Efforts Has MITRE Pursued?
The expanded TempoQuest partnership builds on MITRE’s earlier work in AI-based weather forecasting. In April, MITRE partnered with The Weather Company to integrate its Weather 1K dataset into AI-driven forecasting tools. The high-resolution dataset was developed using weather modeling tools that included TempoQuest’s AceCast, and MITRE is also using it with federal agencies to develop wildfire forecasting capabilities.
The TempoQuest expansion also follows MITRE’s push into advanced computing for national security missions. Through that effort, MITRE and the Department of Energy’s Oak Ridge National Laboratory established the Joint Institute for Computational and Super Intelligence Advantage to develop, test and field AI, quantum computing and other technologies, drawing on resources such as MITRE’s Federal AI Sandbox.
The collaboration also reflects themes MITRE President and CEO Mark Peters raised in an Executive Spotlight interview, where he discussed the push for commercial technologies and a faster path from incubation to implementation for government tools.


