Private Fusion Posts Results, but Nobody Has Put Fusion Power on the Grid

Helion's newsroom posted the news on 13 February 2026. Polaris, the company's seventh-generation prototype, started running at the end of 2024. In January it became the first private fusion machine to burn deuterium-tritium fuel. Helion says it is currently the only one.

That is a real milestone in a field where most private companies have steered clear of tritium, which is radioactive and tightly regulated. Helion says it was also the first company to win regulatory approval to possess and use tritium for demonstrating fusion energy production. The company claims plasma temperatures of 150 million degrees Celsius on Polaris, beating the 100 million degrees Celsius record set by its sixth-generation Trenta machine. The industry treats that level as the threshold for a commercially relevant device. None of this has been independently verified at plant scale, and the fuel mix is not what Helion plans to sell power from.

Two outside researchers are quoted in the announcement. Ryan McBride, an inertial confinement fusion specialist who has worked at Sandia National Laboratories and the University of Michigan, said he reviewed diagnostic data and called evidence of D-T fusion and temperatures above 13 keV, or 150 million degrees Celsius, exciting. Alan Hoffman, who has spent over 40 years on field-reversed configuration plasmas, said he continues to see the technology scaling.

"We believe the surest path to commercializing fusion is building, learning and iterating as quickly as possible," said David Kirtley, co-founder and CEO of Helion.

Note what the D-T result is and is not. Helion says D-T operation is one step in Polaris' testing program. The fuel for commercial operations will be deuterium-helium-3, and the company still has to reach optimal temperatures for it. So the machine that posted the record is not running the fuel Helion intends to sell power from.

Other companies are buying time with hardware

Helion is not alone in announcing firsts. Realta Fusion said on 19 June 2026 that it had demonstrated direct energy conversion of plasma kinetic energy into electricity on the Wisconsin HTS Axisymmetric Mirror device, work done with the University of Wisconsin-Madison. The converter drew multiple amps at around 100 volts, enough to light a few bulbs.

Realta is unusually explicit about the limits. Its own writeup states this was neither a demonstration of net electricity production nor large-scale conversion of fusion-born power, and that most of the converted energy on WHAM came from input power, not fusion. The device runs on deuterium only. The company says it will scale to multi-kilowatt and eventually multi-megawatt capability on future machines.

Thea Energy took a different route: money. On 27 May 2026 it announced a $100 million Series B led by Thomas Tull's US Innovative Technology Fund, with General Innovation Capital Partners, Linse Capital and others participating. The cash goes to magnet manufacturing, including a second facility in northern New Jersey, and to siting and construction of Eos, a large integrated stellarator. Thea says it will pick a site later this year and double its team.

Thea also says it was the first awardee to receive Department of Energy certification of its Helios preconceptual design milestone, and that it has built and operated what it calls the world's first superconducting magnet array capable of producing the fields commercial stellarators need.

The public project is still the measuring stick

Against all this, ITER, the publicly funded international reactor in southern France, carries an estimated cost of $22 billion, according to CNET, which visited the site. The machine's 150-million-degree plasma has to sit next to superconducting magnets kept a few degrees above absolute zero, separated by a thin heat shield. Cracks found in that shield's piping in 2020, plus welding distortions and COVID disruption, pushed the schedule back years and added $5 billion in repair costs.

ITER's chief strategic advisor, Laban Coblentz, told CNET that having China, Russia, the US, Europe, Korea, India and Japan collaborate is "either genius or insane." Javier Artola, a scientist modelling ITER's plasma, framed the project as public knowledge. That openness matters: every problem ITER solves is one private companies do not have to solve themselves.

Read the announcements together and a pattern appears. Private fusion has real hardware, real diagnostics and real money. Helion has a site in Malaga, Washington, where it began building Orion, its first commercial machine, in July 2025, intended to deliver electricity to Microsoft. Thea is talking to offtakers. None of that is the same as a power plant selling electricity. The milestones are steps, and the companies, to their credit, mostly say so.

What is missing from the record is independent confirmation at plant scale. Plasma records come from company diagnostics reviewed by invited experts, not from peer-reviewed grid data. That is normal for an industry this young, and it is why the next few years of claims deserve the same scepticism as the last few.