Weather is our first domain.
It is also the smallest thing Jua will do.

We started with the atmosphere because it is the hardest continuous-physics dataset on Earth. Then we put the product in front of the people whose decisions depend on it.

In production across utilities and trading desks worldwide
Axpo
Switzerland
TotalEnergies
France
Shell
United Kingdom
Enel
Italy
Statkraft
Norway
EnBW
Germany
EDF
France
Hydro-Québec
Canada
Adani Energy
India
Vitol
Netherlands
Origin Energy
Australia
ESB
Ireland
Read the energy-trading deep dive
Why accuracy is worth money

Accuracy in physics is worth money. The atmosphere is just the first proof.

€25M
per gigawatt of wind, per year

A 1 GW wind portfolio produces roughly 3 TWh annually. A 20 pp improvement in forecast accuracy, under typical hedging and imbalance-penalty structures, translates to approximately €25M per year. That is the value of one objective on one domain.

$2.6B
to bring a single new drug to market

Most of that money buys physical experiments that fail. Compressing that loop with learned models of the underlying physics is one of the largest unsolved problems in industry. It is the shape of problem this approach is built for.

60%
of global GDP comes from physical industries

Atmosphere, energy, manufacturing, transport, materials, life sciences. Each is governed by physics. Each is a candidate for a foundation model trained on its own data. Atmosphere is the one we have shipped. The rest is the thesis we are testing.

That is the value of one objective in one domain. There are many.

Beyond the atmosphere

The model does not care what fluid it is looking at.

We trained EPT-2 on the atmosphere because it is the largest continuous record of physics on Earth. The base learned the underlying mechanics, not the weather. Specialize it for an airfoil or a shock wave and the physics carries over. The list below is what the same base extends to, the way a language foundation model extends from text to code. One foundation, many use cases. Not a separate model per industry.

Turbomachinery
Jet engines, gas turbines, compressors. Each design iteration costs weeks of HPC time.
Thermal design
Datacenter cooling, EV battery packs, electronics. The bottleneck on nearly every modern device.
Aerospace
External aerodynamics, transonic flows, re-entry. Wind tunnels are slow and expensive.
Materials
Microstructure evolution, corrosion, crystal growth. Decade-long experimental loops.
Drug discovery
Binding dynamics, protein–protein interaction. A search in continuous physical space.
Robotics
Contact dynamics, deformable materials, fluid–structure interaction. The last mile for autonomy.