Something quietly historic happened last week and almost nobody on Wall Street seemed to notice. For the first time, the Department of Energy gave formal safety clearance for a commercial microreactor to operate at full power. Not a zero-power criticality test. Not a paper exercise. Actual sustained power production. The company behind it is called Radiant, based in Los Angeles, and its Kaleidos unit now sits inside the DOME test bed at Idaho National Laboratory ready for the next phase. Meanwhile the public companies most closely tied to the small modular reactor story have been getting hammered to multi-year lows. I’ve been watching this space for a while, and the contrast feels almost surreal.
The Quiet Breakthrough That Markets Overlooked
Radiant’s Documented Safety Analysis approval covers full-power operation. That is the key difference. Every other design in the current pilot program has only cleared zero-power tests so far. Those earlier milestones proved the physics work. This one clears the path to actually generate electricity for weeks at a time. After the preliminary safety analysis was signed off earlier this year, the new approval moves the project into the Readiness Review stage. Once that is complete, the team can fuel the reactor and begin a structured five-phase test campaign.
The plan is straightforward on paper yet ambitious in practice. First criticality comes first. Then power levels rise step by step. The campaign ends with at least 150 hours of continuous full-power running without any operator intervention. Radiant also intends to operate the same unit for roughly 60 effective full-power days spread over several months. The design and the custom fuel will be essentially identical to what future customers receive. In my view that detail matters more than most headlines admit. It reduces the usual gap between demonstration hardware and commercial product.
Each Kaleidos unit is truck-transportable, produces about one megawatt, needs no cooling water, and can run up to five years between refuelings. Design life is listed at twenty years. Factory production is planned at a large campus in Oak Ridge, Tennessee that will also handle fueling, refueling and spent-fuel storage. Customer sites never have to store spent fuel on their own. That single feature solves a practical and political headache that has slowed nuclear projects for decades.
Why Full-Power Approval Changes the Timeline
Zero-power tests are necessary. They confirm the core physics. But they do not answer the harder questions about heat removal, control systems under load, and long-duration reliability. Full-power operation under continuous monitoring does. Once Radiant completes the DOME campaign successfully, the company will have the most advanced operational data of any U.S. microreactor developer right now. That data becomes the foundation for commercial licensing conversations later.
I’ve found that markets often under-react to regulatory milestones that do not produce immediate revenue. This looks like one of those moments. The approval itself does not generate cash flow next quarter. Yet it removes a major technical and regulatory uncertainty that every competitor still faces. The difference between “we proved it can go critical” and “we proved it can make power for weeks without intervention” is larger than it first appears.
The Commercial Pipeline Already Taking Shape
Radiant is not waiting for the test campaign to finish before lining up customers. The company already holds a binding agreement with the Army through the Defense Innovation Unit worth up to 750 million dollars for fifteen units under the Janus program. That is one of the larger awards the unit has issued. The Air Force selected Radiant earlier to develop and operate a microreactor at Buckley Space Force Base in Colorado, with first commercial units targeted for Buckley and Fort Benning in 2028. More recently the Navy picked the same design for Naval Weapons Station Crane in Indiana. That makes Radiant the only nuclear developer currently working with all three service branches.
On the commercial side a major data-center operator has signed for twenty units. The behind-the-meter angle is obvious. Hyperscalers need reliable power that does not wait years in interconnection queues. A factory-built, truck-delivered reactor that arrives ready to run for five years without water cooling looks attractive on paper. Whether the economics hold once real operating data arrives remains the open question, of course. Still, the pipeline is more concrete than most early-stage nuclear stories can claim.
Factory production is scheduled at the 300,000-square-foot campus in Tennessee. The same site will manage the entire fuel cycle for customers. That vertical approach keeps spent fuel under controlled conditions and simplifies site requirements. In a sector where every extra regulatory layer adds years and millions, the design choices look deliberate.
How Public Nuclear Names Are Trading Right Now
While Radiant advances, the listed companies most associated with the small modular reactor narrative have been punished. One leading name closed recently near levels last seen more than a year earlier after an 86 percent drop from its peak. Another sits even lower on a relative basis. A third has given back roughly three years of gains. October alone saw drops of 25 to 30 percent across several of the pure-play names while the broader market barely moved.
The only relative bright spot has been the large utility that already owns operating reactors and sells power under long-term agreements. Existing cash-flowing assets have held up far better than development-stage stories. That pattern is familiar. Markets pay for megawatts that already exist far more readily than for megawatts promised for the next decade.
Perhaps the most interesting aspect is how quickly sentiment flipped. Earlier this year many of the same stocks were treated as pure plays on the AI power demand theme. Now they trade as if the entire sector faces existential delay. The underlying technology progress has not reversed. The discount rate applied to distant cash flows has simply risen.
Interest Rates and the Brutal Math of Distant Cash Flows
The sharpest leg lower arrived after a hotter-than-expected inflation reading pushed the ten-year Treasury yield higher. For companies whose first meaningful revenue sits years away, every basis-point increase on the long end cuts the present value of future earnings. Valuation models that discount 2034 or 2035 cash flows become sensitive to small changes in the risk-free rate assumption.
Analysts have been adjusting targets accordingly. Some still maintain constructive ratings while cutting price objectives. Others have moved more aggressively to underweight or sell recommendations, citing higher rates, policy uncertainty and rising costs. Street targets now span a wide range, which itself tells a story. When the gap between the most optimistic and most cautious forecasts becomes that large, it usually means the market is still searching for a durable valuation framework.
Dilution adds another layer of pressure. Share counts have risen meaningfully at several of the pure plays. Cash balances look healthy for now, yet the projected spend through the end of the decade is substantial. One company recently launched a sizable at-the-market equity program. Capital markets remain open, but each raise increases the hurdle future projects must clear to deliver attractive equity returns.
Where the Listed Developers Actually Stand
Regulatory clocks for the public names still stretch measured in years rather than months. One company achieved criticality with a test reactor and received preliminary safety analysis approval for its first commercial unit, yet commercial operation remains targeted for 2028. Another saw its construction permit application accepted, with a final safety evaluation currently scheduled around late 2027. The only design that already holds full design certification still lacks the volume of binding commercial orders that would shift the narrative.
Radiant’s current schedule aims to begin delivering units to the military in 2028 and to have fifteen units in Army service by 2030. If the full-power campaign proceeds as planned, the first company to demonstrate sustained commercial-scale microreactor operation in the United States could well be a private firm. Public investors cannot buy shares in that progress directly. That structural gap helps explain part of the current valuation divergence.
I’ve noticed that successful nuclear stories historically required both technical proof and a clear path to capital recovery. The technical side is advancing faster than many expected two years ago. The capital recovery side remains more uncertain, especially while long-term rates stay elevated.
The Broader Nuclear Demand Backdrop Has Not Disappeared
Utilities and large technology firms continue to seek every available megawatt of firm, carbon-free power. Uprate projects at existing plants, restart discussions for previously closed reactors, and long-term power purchase agreements keep appearing. The interconnection queue in several regions has grown long enough that some data-center developers are exploring behind-the-meter solutions more seriously than before.
Small modular and microreactor designs sit at the intersection of that demand and the practical constraints of the grid. They promise shorter construction cycles, factory fabrication, and the ability to place generation closer to load. Whether those promises survive contact with real-world cost and schedule data is still being tested. Radiant’s full-power campaign is one of the clearer near-term tests of that proposition.
The military interest is particularly noteworthy. Bases need resilient power that can operate independently of the commercial grid. A truck-transportable unit that arrives, connects, and runs for years without large cooling infrastructure fits that requirement in theory. Multiple service branches placing orders or selecting the same design suggests the operational case is gaining internal support.
Positioning and Sentiment Have Reached Extremes
At one point earlier this autumn some institutional research noted that positioning in uranium-related equities had fallen to multi-year lows even while physical and policy indicators remained constructive. Subsequent price action showed that positioning can always get lighter. The nuclear equity complex has underperformed the broader market by a meaningful margin over recent months while short interest rose in several names.
That kind of extreme often creates the conditions for eventual mean reversion, but timing remains difficult. Rate relief would help. Clearer evidence of commercial offtake would help more. Successful full-power demonstration by any microreactor developer would provide tangible proof that the technology path is real. Radiant is currently closest to delivering that proof.
In my experience, markets can stay disconnected from fundamental progress longer than most observers expect. The current episode looks like another example. Engineers keep clearing milestones. Equity valuations keep reflecting higher discount rates and capital needs. Both observations can be true at the same time.
What Successful Full-Power Operation Would Actually Prove
The 150-hour continuous run without operator intervention is more than a marketing claim. It tests thermal management, control system stability, and passive safety features under real load. Extending that into 60 effective full-power days adds information about material performance and fuel behavior over longer periods. Those data sets become the reference points regulators and future customers will examine.
Because the unit uses the same basic design and fuel intended for commercial deployment, the test campaign is not purely academic. Success would move the conversation from “can it work” to “how quickly can it be manufactured and deployed at scale.” That shift has been missing from most public microreactor stories so far.
Failure or significant delay would of course reinforce the skepticism currently priced into the sector. The stakes for the broader narrative are therefore higher than a single company’s project. The first successful full-power commercial microreactor demonstration in the United States will set a benchmark against which every other design will be measured.
Capital Intensity Remains the Central Challenge
Even with technical progress, the capital required to reach commercial scale is large. Factory construction, fuel fabrication capacity, first-of-a-kind engineering, and regulatory engagement all consume cash before meaningful revenue arrives. Public companies must continually return to equity markets or secure project financing under conditions that have grown less forgiving.
Private companies face similar absolute capital needs but different investor bases and different disclosure requirements. That difference can allow them to advance more quietly through early stages. Radiant’s current position illustrates the point. The most significant microreactor milestone of the year belongs to a firm whose equity is not available to public market investors.
Over time the two worlds usually converge. Successful private developers eventually seek larger capital pools or strategic partnerships that bring them into the public eye. Until then the valuation gap can persist. Investors focused solely on listed names risk missing the actual pace of technical advancement occurring outside their investable universe.
Policy Support and the Regulatory Machine
The current administration has emphasized accelerating advanced nuclear deployment. The pilot program at Idaho National Laboratory and the relatively rapid movement of several designs through early safety reviews reflect that priority. Full-power approval for Radiant fits inside that broader effort. Officials have described the step as consistent with the goal of restoring American leadership in nuclear technology.
Policy support alone does not guarantee commercial success. It can, however, shorten the timeline between design completion and first operation. In a capital-intensive industry, time is almost as important as money. Every year of delay compounds financing costs and allows competing technologies to claim more of the near-term power demand.
The regulatory path for microreactors is still being written in real time. Early successful demonstrations will influence how subsequent applications are reviewed. That feedback loop makes the current test campaigns more consequential than they might appear from the outside.
The Data Center Angle and Behind-the-Meter Demand
Large technology firms continue to sign long-term power agreements and explore on-site generation options. Grid constraints in key regions have made the traditional utility interconnection process less attractive for time-sensitive projects. A factory-built microreactor that can be delivered and brought online on a more predictable schedule offers one potential answer.
Whether the cost per megawatt-hour ultimately competes with other firm power options will determine how widely the technology spreads. Early military and data-center contracts provide the first real offtake data points. Those contracts also give manufacturers the volume needed to drive down production costs over successive units.
I’ve found that the most durable energy technology stories usually combine a clear operational need with a path to cost reduction through manufacturing scale. Microreactors are still early on that curve. The next few years of operating experience will show how steep the curve can become.
Valuation Frameworks Still Struggle with Long-Duration Assets
Discounted cash flow models remain the primary tool analysts use for these companies. Small changes in the terminal growth rate, the weighted average cost of capital, or the assumed commercial online date produce large swings in present value. When the risk-free rate itself is moving, the models become even more volatile.
Some research houses have attempted to anchor valuations to comparable infrastructure or utility multiples once projects reach commercial operation. That approach only becomes useful after first power is delivered. Until then the sector trades more on narrative, positioning, and rate expectations than on traditional earnings multiples.
The current environment rewards companies that already generate cash and punishes those that must raise it. That is rational in a higher-rate world. It also means that genuine technical progress can go under-appreciated for extended periods. Radiant’s approval is a reminder of that dynamic.
Looking Ahead to the Next Set of Catalysts
Several public companies are scheduled to report quarterly results in the coming days. Consensus expectations for revenue at the pure development names remain modest, as one would expect at this stage. Guidance on cash burn, project timelines, and any new commercial discussions will matter more than the headline numbers.
On the private side the Readiness Review for Kaleidos and the subsequent start of the full-power campaign will be watched closely by anyone following the sector. Successful completion of the 150-hour continuous run would represent the clearest operational proof point yet for U.S. microreactor technology.
Broader market conditions will continue to influence equity performance. Any sustained move lower in long-term yields would ease pressure on distant cash-flow valuations. Conversely, further rate increases or renewed inflation concerns would likely keep the group under pressure regardless of individual project news.
Balancing Technical Progress Against Market Reality
The engineers and the market are both right in their own way. The technical case for modular nuclear has strengthened. Regulatory processes that once seemed immovable have begun to produce tangible approvals on faster timelines. Military and commercial interest is real. At the same time, the equity market is correctly applying higher discount rates to cash flows that still sit years in the future and that will require additional capital raises to reach.
That tension is unlikely to resolve quickly. Full-power demonstration data will help on the technical side. Lower rates or clearer offtake contracts will help on the valuation side. Until both improve together, the disconnect can persist.
There is something almost poetic about a reactor that must prove itself with 150 hours of uninterrupted full power. The nuclear equity complex may need a similar stretch of supportive conditions from the bond market before sentiment turns lasting. In the meantime the most important progress continues to occur largely outside the public markets. That fact alone makes the current moment worth watching closely.
The story of advanced nuclear in the United States is no longer purely theoretical. Hardware is moving. Approvals are being granted. Contracts are being signed. Whether public investors will be able to participate profitably in the next phase depends as much on the path of interest rates and capital markets as on the physics of fission. For now those two forces remain in open conflict, and the scoreboard favors the bond market.
