The sodium cooled test reactor reached zero power criticality at Idaho National Laboratory on June 4, 2026, validating core physics rather than a path to commercial power.
Antares Nuclear's Mark-0 test reactor reached initial criticality on June 4, 2026, at Idaho National Laboratory, according to New Atlas. New Atlas reports it is the first privately developed, non-light-water reactor to go critical in the United States in more than 40 years.
The word "critical" does a lot of work in that sentence. The Mark-0 reached what engineers call initial, or zero-power fueled, criticality: the minimum configuration in which a self-sustaining nuclear chain reaction can be maintained, with power levels too low to produce usable electricity. The reactor is not connected to a grid. The milestone is about the physics of the core, not about megawatts.
That distinction matters because the technology is the point. The Mark-0 is sodium-cooled, modular, and designed to run at higher temperatures than the light-water reactors that make up nearly the entire US commercial fleet. Sodium transfers heat far more efficiently than water, which lets a reactor operate at higher temperatures and lower pressures. That opens the door to more efficient electricity generation and to industrial heat uses, but it also brings engineering costs: sodium burns on contact with air or water and becomes opaque, complicating inspection and maintenance. Light-water reactors have decades of operational data and a deeply established supply chain. Sodium-cooled designs have had neither in the United States for a long time.
The Mark-0 is the test platform for the design. The stated goal of the test campaign is to validate computational physics models, core geometry, control-rod performance, and initial neutronic behavior. In other words, the team is checking whether the simulation on a screen matches the physics in a tank of liquid sodium, before committing to a higher-power or production unit. The "Mark-0" name itself signals a development ladder.
New Atlas reports the milestone sits inside the Department of Energy's Reactor Pilot Program, a pathway intended to move advanced reactor designs from lab to demonstration faster than the conventional Nuclear Regulatory Commission commercial licensing route. The question that follows is whether the program produces a template other private developers can use, or whether it amounts to a well-timed exception for a well-prepared first mover. Critics of advanced nuclear projects point to a long history of cost overruns and missed timelines. Supporters argue that modular construction and a faster regulatory path are the changes needed to break that pattern.
The next test is whether the model survives contact with the real thing. The bigger question is whether the unit above the Mark-0 on the development ladder ever gets built.