AMPERA | Power Now. Nuclear Next.

AMPERA nuclear power system: reactor, heat exchanger and power conversion modules in a studio

PowerNOW · commercially available Q1 2028

Factory-built sCO₂ power on gas today. A sealed thorium core on the same platform tomorrow. No stranded asset.

The constraint

Compute is scaling faster than the grid can connect it. So the power has to come to the load.

U.S. electricity usage and generation, TWh per year. Curve shape simplified; 2035 values as published in AMPERA Master Financial Projections.

2035 projected shortfall

1,574

TWh, about 180 GW continuous

Usage grows 64%, generation 26%. AI load alone adds 1,100 TWh, slightly more than all projected U.S. generation growth.

The nuclear module. In the round.

0°Rotation

01 / Nuclear reactor module

Thorium core

A 3D-printed, solid-state thorium core inside a shielded sphere. No refueling for a target of up to 30 years.

02 / Heat exchanger module

Heat into the loop

Core heat moves into a closed supercritical CO₂ circuit. No water, no steam plant.

03 / Power conversion module

Brayton-cycle power

Compact sCO₂ turbomachinery and generator, targeting more than 50% efficiency.

04 / Containerized

Built to ship

Factory-built modules at 7.5 or 15 MWe. Delivered, connected, monitored.

One platform

Same power block. Swap the heat source.

Deploy PowerNOW on gas today. When the nuclear module is licensed, it replaces the thermal module. Same power block, same site, same interconnect.

AMP-G7.5 PowerNOW system: burner thermal module, power module and heat rejection module — **Power now**Flex fuel: gas, LNG, CNG, diesel, hydrogen

AMPERA nuclear system: reactor module, heat exchanger module and power conversion module — **Nuclear next**Sealed thorium core

Where it deploys

Firm power where the grid can't reach.

Containerized, factory-built, rapidly deployable. One product line for four markets.

AMPERA nuclear system inside a data center hall

Data centers

Firm, efficient baseload for edge and hyperscale compute, up to multi-GW campuses.

AMPERA modules at a forward military base

Defense

Deployable microgrids for field operations and critical infrastructure.

Industrial

Reliable on-site power for heavy industry and remote, off-grid sites.

AMPERA modules on a port quay beside a ship

Maritime

Dockside, floating barge and vessel propulsion power.

Nuclear next

A reactor you
never refuel.

The first full-scale 3D-printed nuclear reactor module: solid state, built around a proprietary gyroid structure with microchannels. It runs subcritical, driven by an external neutron source, so it switches on and off instantly and is passively safe.

Output: 7.5 / 15 MWe
Efficiency target: >50%
Life, no refueling: Target up to 30 years
Architecture: Subcritical, accelerator driven
Fuel: Thorium TRISO, liquid-metal jet printed

The technology

4×more abundant than uranium

1000×smaller waste footprint

100×lower cost than HALEU

The power block · sPower

Reinventing power conversion.

Closed-loop supercritical CO₂ replaces the steam Rankine cycle with a high-temperature Brayton cycle. Compact, water-free, quieter, and agnostic to its heat source. Future-proofed for fission and fusion.

System efficiency: 50–60%
Size reduction: 50%, via additive manufacturing and nickel superalloys
Heritage: 20 years of R&D, thousands of operating hours

sPower: supercritical CO2 power block — sPower conversion module

AMPERA power conversion module: sCO2 turbomachinery and generator behind a glass panel — sPower conversion module

Factory production

Built on a line. Shipped on a truck.

Rapid scaling comes from the factory, not the job site. Modules are fabricated, integrated and tested under one roof, then lifted onto a trailer.

01/ 11

Receiving dock with forklifts unloading pallets of material — 01Freight in

CNC machining cells and fabrication stations along a factory aisle — 02Fabrication and machining

Additive manufacturing cells with metal printers and inspection benches — 03Additive manufacturing

Overhead view of four parallel PowerNOW assembly lines — 04Four PowerNOW lines

Finished module carried down the line transport aisle on an autonomous transporter — 05Line transport aisle

Heat rejection modules being integrated at assembly stations — 06Heat rejection integration

Technicians wiring electrical and instrumentation and control cabinets — 07Electrical and I&C integration

Quality inspectors performing non-destructive examination on a module — 08Quality and NDE hold

Control room operators monitoring three modules under test — 09Connected system test

Finished systems staged in rows awaiting shipment — 10Finished system staging

Overhead crane lifting a finished module onto a flatbed trailer — 11Crane lift to trailer

Concept renderings

Economics, target

Redefining nuclear economics.

A target long-term LCOE of about $20/MWh, from factory-built reactors and a sovereign thorium fuel strategy.

Capital cost$M / MWe

AMPERA$1.7

Solar PV + storage$1.4

Gas combined cycle$1.5

Wind + storage$2.1

Nuclear, U.S.$6.0

Fusion$15.0

Levelized cost$ / MWh

AMPERA$20

Wind + storage$44

Gas combined cycle$48

Solar PV + storage$50

Nuclear, U.S.$141

Fusion$200

Sources: non-company benchmarks from publicly available third-party data (Lazard LCOE+ 2025 low-end estimates, U.S. EIA AEO2025, DOE and OECD/NEA advanced nuclear and SMR studies, peer-reviewed fusion economics). AMPERA values are internal target economics assuming scaled factory-built deployment and are not demonstrated commercial operating data. LCOE is the average cost to generate 1 MWh over a project's life, including capital, operations, maintenance and fuel. Capital cost is the estimated $M invested per MW of electricity.

Affiliations