General Fusion Says Its LM26 Machine Heated Plasma Electrons Past 1 keV

General Fusion said on October 8, 2026 that its LM26 machine had heated plasma electrons past 1 keV, a long-standing benchmark in fusion research. The Canadian company announced the result jointly with the UK Atomic Energy Authority (UKAEA), Britain's national fusion laboratory. UKAEA said the reading was 1.1 keV, or about 12.6 million degrees Celsius, taken just before peak compression. The paper behind it has been submitted for peer review but not yet published in a journal. The site's fusion milestones tracker logs it as a reported result, not yet peer reviewed.

What did LM26 measure?

Plasma physicists often give temperature in electronvolts. One kiloelectronvolt, or keV, is 1,000 electronvolts.

The headline reading came from Thomson scattering. UKAEA explains that the method shines a laser through the plasma and analyzes how the light scatters. That shows how hot the plasma is without disturbing it.

The Thomson scattering paper gives the reading as 1,090 electronvolts, plus or minus 40. It came from a compression shot called LMC-12. The paper says the system can measure at 8 times and 2 points during a compression. Plasma density rose about tenfold as the plasma was squeezed.

General Fusion's own release adds figures that UKAEA's release does not carry. A second instrument, which the company calls an absolute extreme ultraviolet (AXUV) diagnostic, read about 1.2 keV, or 13.7 million degrees Celsius. The company says it measured at peak compression, a point the Thomson system could not reach. The company also reported ion heating to about 0.46 keV. Ions are the atomic nuclei that actually fuse.

Those figures come from a second paper, also submitted for peer review. Its summary gives a peak electron temperature of 1,180 electronvolts, plus or minus 65. It puts the ion temperature at 459 electronvolts, plus or minus 33, inferred from the plasma's neutron output. The paper says that margin covers counting statistics only. It also says shrinking the liner's radius by a factor of 2.75 raised the electron temperature more than fivefold.

Who checked the result?

UKAEA is a partner in the work, not an outside referee. Its specialists designed and commissioned the Thomson scattering system together with General Fusion's. UKAEA built a key part, a polychromator that analyzes the scattered light, at its Diagnostic Innovation Centre of Excellence at the Culham Campus. It also co-wrote the paper. The 1.2 keV reading and the ion figure come only from General Fusion. Every author of the second paper lists the company as their affiliation. That paper thanks UKAEA for help with neutron simulations, but UKAEA's release does not mention either figure.

The Globe and Mail reported that the company partnered with UKAEA to get independent confirmation of its 1 keV measurement. Nuclear Newswire, published by the American Nuclear Society, reported both readings as company statements. Neither report quoted an outside scientist assessing the data.

UKAEA chief executive Dennis Whyte said hot electrons are a necessary step for proving fusion confinement concepts. He said UKAEA looks forward to the outcome of peer review and to the paper's publication in the journal Review of Scientific Instruments. As of October 15, 2026, General Fusion's website still listed both papers as submitted for peer review.

Why does 1 keV matter?

Fusion happens when nuclei collide, so the ion temperature is what drives reactions. The Globe and Mail noted that the 1 keV figure refers to electrons, not the hydrogen nuclei. General Fusion's chief strategy officer, Megan Wilson, said the nuclei also showed "meaningful heating," according to the paper.

Hot electrons still matter, because electrons can cool a plasma quickly. Zap Energy called passing 1 keV a key hurdle for that reason when it reported its own result in 2024. Its FuZE device measured 1 to 3 keV, in a paper published in Physical Review Letters. The Department of Energy's Scott Hsu said then that only a handful of fusion concepts had reached a 1 keV electron temperature.

The threshold has a long history. Nuclear Newswire notes that 1 keV was first reported on the Soviet T3 tokamak, announced in 1968. UK physicists verified it with Thomson scattering the following year. The newswire lists Tokamak Energy, Helion and Zap Energy among the few private companies to report the milestone before General Fusion.

General Fusion's claim of a first is narrower. It says LM26 is the first machine to reach 1 keV with low-speed compression.

How does magnetized target fusion work?

General Fusion's approach is called magnetized target fusion. It holds a plasma in magnetic fields, then squeezes it with a collapsing metal wall. The design avoids superconducting magnets and high-powered lasers.

In LM26, the plasma sits inside a solid lithium liner that starts 1.76 meters across. Coils implode the liner around it, according to the second paper. The squeeze takes milliseconds. Other compression methods work in microseconds or nanoseconds, the company says. It argues that the slower pace means less power, less complexity and lower cost.

A power plant would work differently. The company plans a liquid metal wall that would shield the machine from neutrons, breed fuel and capture heat to make steam. The Globe and Mail reported that such a reactor would compress plasma in a repeating cycle, not single shots like LM26.

What comes next for LM26?

General Fusion says upgrades are underway to reach higher compression ratios. Its next target is 10 keV, which chief executive Greg Twinney put at roughly 100 million degrees Celsius. After that comes the Lawson criterion. That is the mix of temperature, density and confinement time needed for net fusion energy in the plasma.

Twinney wrote that the company targets the Lawson criterion by the end of 2028. It aims to begin operating a first-of-a-kind plant around 2035. Both are company targets.

The schedule has slipped before. When General Fusion announced LM26 in August 2023, it aimed for over 100 million degrees Celsius by 2025, Nuclear Newswire reported. It also aimed to progress toward scientific breakeven by 2026. In June 2026, the company reported 0.72 keV and named 1 keV as its next goal.

Money shaped the timeline. TechCrunch reported that General Fusion laid off at least 25% of its staff in May 2025. It began trading on Nasdaq in July 2026. TechCrunch said funding problems had likely pushed breakeven to 2028 or later. On the day of the 1 keV announcement, its shares closed up more than 14% at US$8.71, The Globe and Mail reported. Twinney said the capital raised as the company went public "funds us through the next couple of milestones," the paper reported.

Several questions remain open. Peer review could confirm or qualify the readings. The October releases give no date for reaching 10 keV. General Atomics is advising on the measuring system for that step. For where other fusion efforts stand, see Fusion Power: How Close Are We, Really?