Engineers in Everett ran tritium through a 60 foot fusion prototype and pushed the plasma to 150 million degrees, and it took five separate detectors to confirm the reaction had happened at all

A low industrial building near the Washington coast, in a city better known for assembling airliners.

Inside it, a tube about 60 feet long, wrapped in copper coils and banded with steel.

Nothing about it looks like a power station. There is no turbine and no cooling tower.

Asked about scientific breakeven he was direct about the priority, saying the company focuses on the electricity piece rather than the pure scientific milestones.

The whole event lasts under a millisecond.

In that fraction of a second the gas inside reaches ten times the temperature at the center of the sun.

Then everything shuts off and the machine waits.

How the machine gets there and takes the energy back

Most people picture fusion as a doughnut holding a plasma in place for as long as possible. This is the opposite approach.

Two separate clouds of ionized gas are formed at opposite ends of the tube, then accelerated toward each other by magnets firing in sequence along its length.

They collide in the middle and merge into a single spinning ball, arriving at somewhere between 10 and 20 million degrees.

That is already extraordinary and it is nowhere near enough.

A second and much harder magnetic squeeze then compresses the merged ball the rest of the way.

What happens next is the part that makes this design unusual. The expanding plasma pushes back against the magnetic field that squeezed it, and that push induces a current in the same coils, so electricity comes straight back out of the machine without ever passing through steam or a turbine.

What changes when you put tritium in it

Most laboratories run on deuterium, which is stable, safe to handle and abundant in seawater.

Tritium is a different proposition. It is rare, mildly radioactive, and it leaks through materials that hold almost anything else.

Handling it demands containment, monitoring and a regulator’s permission across the entire facility.

So running it is not really a physics achievement. It is an engineering and licensing one, and this is the first private machine cleared to possess and use it for a fusion demonstration.

The payoff is that deuterium and tritium react far more readily than deuterium alone.

They also throw off a very specific signature, a neutron carrying 14.1 million electron volts, which is what makes the reaction possible to prove.

The five instruments that had to agree

The temperature reached 150 million degrees Celsius in January of 2026, which physicists write as about 13 kilo electron volts.

Proving that this fusion prototype had actually fused anything took more than one instrument.

Organic scintillators saturated on the first pulses, running roughly ten times higher than comparable deuterium runs. A diamond detector array separated the 14.1 million volt neutrons from the weaker ones a deuterium reaction makes.

A fused silica detector built in house responds only to the high energy neutrons, so any signal from it is conclusive on its own.

Small copper discs were exposed inside the machine and then read by gamma ray spectroscopy, and they came back activated.

A fifth instrument caught the alpha particles directly, at the exact energy the reaction produces.

What none of that establishes

No net energy gain was demonstrated and none was claimed.

The record is the highest for a privately built machine, not for fusion in general, and national laboratories have reached comparable numbers.

Peer reviewed publication is also thin. The chief executive has said the team is preparing papers on the diagnostic methods, which means outside verification is still pending.

Asked about scientific breakeven he was direct about the priority, saying the company focuses on the electricity piece rather than the pure scientific milestones.

His description of the result was that they saw the fusion power output increase dramatically as expected in the form of heat, and the measurement write up backs that narrower claim.

Taking power straight out of a plasma is not unique to this company either, and other groups are extracting electricity from magnetic devices directly.

The contract sitting downstream of the laboratory

What makes this consequential is the commitment already in place.

The company broke ground in the summer of 2025 on a commercial plant in eastern Washington, built to sell electricity to a major technology customer starting in 2028. A separate agreement targets a 500 megawatt plant for steelmaking around 2030.

That turns a research program into a delivery obligation.

The company was founded in 2013 and is on its seventh prototype, each one built, tested and retired in sequence, which is closer to how aerospace iterates than to how government fusion programs work.

Two timelines are now running at once, one in the laboratory and one on a construction site.

The gap between them is where the hard problems live, and the record set this year says nothing about how quickly a 2028 deadline can be met.

What was measured in Everett is real. What is promised for two years from now is an entirely different claim.