DOE: four advanced reactors hit criticality in 2026

The U.S. Department of Energy says four advanced test reactors reached criticality in 2026, beating a presidential target of three by July 4. In a July 25 fact sheet, DOE presents the tally as proof the Reactor Pilot Program is producing repeatable hardware milestones, not only project announcements.

Key facts

• DOE says it exceeded a July 4, 2026 goal of three advanced test-reactor criticalities by recording four.
• The four named milestones are Antares Mark-0 (June 4), Valar Ward 250 (June 18), Deployable Energy Unity (June 30) and Aalo-X (July 4).
• Mark-0's zero-power demonstration at Idaho National Laboratory was described by DOE as the first privately developed non-light-water reactor criticality in the U.S. in more than 40 years.
• The target originated in a May 2025 executive order on reforming nuclear reactor testing at DOE.
• DOE launched the Reactor Pilot Program in June 2025 and announced initial project selections in August 2025.

What happened

DOE's fact sheet is a victory-lap document with a hard center of gravity: named reactors, named dates, named sites. According to the department, Antares Nuclear's Mark-0 reached criticality at INL on June 4, 2026 in a zero-power fueled demonstration. Valar Atomics' Ward 250 followed on June 18 at the Utah San Rafael Energy Lab in Emery County. Deployable Energy's Unity completed a criticality demonstration at INL on June 30. Aalo Atomics' Aalo-X finished a zero-power fueled criticality demonstration at INL on July 4.

That sequence matters because the public goal was not "progress toward criticality." It was criticality itself, three times, by Independence Day. DOE says it cleared the bar and then kept going. The department also folds the tally into a broader administration narrative on licensing speed, fuel supply chains and plant restarts, but the countable claim is the four test-reactor criticalities.

Why it matters

Advanced-reactor coverage is full of renderings. Criticality is not a rendering. Even a zero-power demonstration is a physics and safety-case event: operators load fuel, establish procedures, and show the machine behaves as the models said it would under tightly controlled conditions.

For the U.S. cohort, the signal is repetition. One private non-LWR criticality can be dismissed as a one-off. Four in a single season, under a named pilot program, is a cadence claim. It also reframes nearby milestones NNN has already covered — including Oklo's Groves startup authorization and the DOME microreactor test bed — as parts of the same pipeline from paper reactor to controlled experiment.

Background

The Reactor Pilot Program was DOE's answer to an executive-order deadline: accelerate advanced-reactor testing inside the department's complex and get multiple machines to criticality fast. Initial selections in August 2025 put a roster of private developers into that funnel. Not every selected company is in the four-criticality list, and not every criticality is a power-producing plant. Mark-0 and Aalo-X are explicitly described as zero-power demonstrations establishing baselines for later electricity-producing units.

That distinction is the adult version of the headline. The U.S. has not suddenly opened four commercial advanced reactors. It has, if DOE's account holds, run four controlled first-criticality events on an accelerated federal testing pathway. Historically, that is still a sharp break from the long drought in privately developed non-LWR critical experiments.

What's next

Watch three follow-through tests. First, which of the four programs convert zero-power success into electricity-producing demonstrations in 2027 and beyond. Second, whether DOE's criticality cadence continues after the July 4 political deadline loses force. Third, how quickly fuel supply — especially HALEU and TRISO fabrication — keeps up with a pipeline that is finally generating experimental throughput instead of only press releases.

The fact sheet also places the criticality streak inside a wider financing and restart campaign. DOE highlights Office of Energy Dominance Financing support for long-lead supply-chain items aimed at ten new large commercial reactors, plus restart work at plants such as Palisades in Michigan and the Crane project in Pennsylvania. Those are not the same as a microreactor zero-power test, but they are part of the same industrial argument: the United States is trying to move nuclear projects through multiple gates at once — testing, fuel, restarts and new large builds — rather than waiting for a single flagship design to de-risk the sector.

For NNN readers, the editorial test is simple. If the next six to twelve months produce only more fact sheets, the July criticality claim becomes a closed political story. If it produces instrumented operating histories, fuel contracts and follow-on power demonstrations, then July 2026 really was an inflection point in the advanced-reactor experimental pipeline.