Fervo Energy is teaming up with Nvidia and Pacific Northwest National Laboratory (PNNL) to develop EGS-Twin, a digital twin platform designed for enhanced geothermal systems. Think of it as a virtual replica of underground geothermal reservoirs, powered by AI, that lets engineers simulate and optimize energy extraction before drilling a single additional well.
The collaboration sits at a surprisingly natural intersection. Nvidia brings the GPU horsepower and AI frameworks. PNNL, a US Department of Energy national lab, brings deep subsurface science expertise. And Fervo brings the real-world geothermal operations to ground-truth the whole thing.
What a digital twin means for geothermal
EGS involves injecting fluid into hot rock deep underground to create artificial reservoirs, then circulating water through fractures to extract heat. The subsurface is messy, heterogeneous, and difficult to image precisely. Building an accurate digital twin of that environment requires serious computational muscle and sophisticated physics modeling.
Nvidia has already been collaborating with DOE national labs on AI-driven projects across the energy sector, including nuclear energy applications. Nvidia is also a founding member of the Enhanced Geothermal Systems Deployment Coalition, which counts Fervo among its members.
PNNL, based in Richland, Washington, has decades of experience in subsurface science, computational modeling, and energy systems research.
Fervo’s rapid ascent
Fervo’s Project Red pilot began delivering 3.5 MW of enhanced geothermal power in November 2023, supplying clean energy to Google data centers.
Its Cape Station project in Utah is targeting 100 MW by October 2026, with plans to eventually reach 500 MW by 2028.
Fervo raised $462 million in a funding round in December 2025, bringing its total funding since 2017 to roughly $1.5 billion. Then in May 2026, the company completed an IPO on the Nasdaq, raising approximately $1.89 billion.
What this means for investors
Geothermal’s biggest obstacle has always been subsurface uncertainty. Wells are expensive. Drilling into the wrong spot is very expensive. If EGS-Twin can reduce that uncertainty through better simulation, it directly attacks the economic bottleneck that has kept geothermal from scaling. Lower risk per well means lower capital costs, which means more projects get financed, which means faster deployment.
Fervo has actual megawatts flowing to actual customers, a freshly public stock, and now AI partnerships with some of the most well-resourced organizations on the planet. If Fervo’s Cape Station hits its 100 MW target on schedule later this year, the question of whether EGS can compete with solar and wind on a levelized cost basis may start coming into focus.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.

1 hour ago
1
















English (US) ·