MITRE, TempoQuest Expand GPU Weather Forecasting Push
The partners will adapt the MPAS weather model to run on GPUs for faster federal access to forecast data.
Updated on Nov. 19, 2026 in Artificial Intelligence

MITRE and TempoQuest expanded their partnership to apply GPU-based weather prediction to federal decision-making, announced Wednesday. The partners will begin by adapting the Model for Prediction Across Scales (MPAS) to run on GPUs.
Why it matters
MPAS's heavy computing demands can slow its simulations, so moving it onto GPUs could speed the forecasts agencies rely on for time-sensitive decisions. The effort ties into federal forecasting modernization, including NOAA's incorporation of MPAS into its next-generation systems.
The Weather 1K AI training dataset is roughly 7 petabytes in size and was built with TempoQuest's AceCAST platform. Performance gains from the GPU adaptation have not been quantified in the announcement.
The players
MITRE
MITRE is a nonprofit organization that operates federally funded research and development centers for U.S. government agencies.
TempoQuest
TempoQuest is a company that developed the AceCAST platform used to build the Weather 1K dataset.
Oak Ridge National Laboratory
Oak Ridge National Laboratory is a U.S. Department of Energy science facility focused on computing and energy research.
The details
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.
Timeline
April: MITRE partnered with The Weather Company to integrate Weather 1K into AI-driven forecasting tools.
Wednesday, November 18, 2026: MITRE announced the expanded partnership.
Tech Race
The move extends the MPAS framework that NOAA is building into its next-generation forecasting systems, adding GPU acceleration to a model federal agencies are already adopting. It positions MITRE and TempoQuest in a federal forecasting effort that is increasingly computation-bound.
Faster GPU-based forecasts could give federal agencies quicker access to weather data for time-sensitive decisions. The practical effect for the public depends on how quickly agencies adopt the adapted model.
The takeaway
Faster, GPU-accelerated weather models could shorten the time between data collection and actionable forecasts for federal decision-makers. Watch whether agencies move the adapted MPAS into operational use.
Further reading
For more on the tools reshaping forecasting, see Artificial Intelligence coverage.
Source note: This article includes information reported by ExecutiveBiz.






