Nuclear Stocks: What Is Inside the Nuclear and SMR Theme

The theme applies one test: does revenue or funded development depend materially on nuclear power? In practice that means operating reactors, developing small modular reactors, mining and supplying nuclear fuel, or building the reactor components, turbines, and engineering the fleet runs on.

The word doing the work is materially. A very large number of companies touch nuclear somewhere. A regulated utility with reactors inside a much bigger rate-based fleet is not a nuclear stock in any useful sense, because what moves it is a rate case, not reactor economics. Drop that requirement and the theme quietly becomes a list of utilities and industrials with a nuclear anecdote attached, which is the failure mode of most thematic screens.

The second structural choice is that the theme spans layers rather than picking one. Owning only SMR developers is a bet that a technology commercialises on a particular schedule. Owning only operators is a bet on power prices with a nuclear label. Holding both means one part of the roster has revenue you can read off a financial statement while another part is an option on the industry's cost structure changing, and those two do not fail at the same time. For the general idea, see thematic investing.

Every reactor that will generate revenue this year is already built, and the companies that run those reactors are the only part of the theme with a nuclear business you can measure from current financial statements. This layer sells electricity from an existing fleet, and what changed its standing was demand rather than technology: data centers and AI training campuses need firm, round-the-clock, carbon-free power at a scale few other sources deliver, which turned reactors that once looked like legacy assets into contracted capacity. This is the near end of the theme's time horizon, and it is the reason the roster is not simply a collection of pre-revenue startups.

How this layer relates to the rest. The operating fleet is what makes the rest of the theme legible. It proves there is a buyer willing to pay a premium for firm clean power, which is the demand signal the SMR developers are ultimately underwriting, and it is the fuel layer's actual customer today. If contracted power demand cooled, the developers would lose the commercial argument for their designs long before they lost the technology.

A reactor is only investable if the fuel behind it is. Uranium is mined, converted, and enriched into reactor-grade fuel long before it produces a kilowatt-hour, and utilities buy it years ahead through long-term contracts, which is why this layer runs on its own clock: slower and further out than the operators, but far closer to real revenue than the developers. It also has a property no other layer has, which is that it does not care which reactor wins. Fuel demand tracks the number of reactors running and planned, not the fortunes of any single design.

How this layer relates to the rest. This layer sits between the two ends of the theme and connects them. It supplies the operating fleet today and would supply the SMR fleet if one gets built, so it is the one part of the roster that participates in both the near-term and the long-term version of the thesis. It is also the layer whose price is set outside the theme entirely, by a small global commodity market rather than by power contracts or reactor licensing.

Small modular reactors are the part of the theme with no current revenue and the largest possible payoff. The design premise is a break from how nuclear has been built: instead of a one-off, roughly gigawatt-scale plant assembled on site over many years with severe cost and schedule risk, an SMR is meant to be smaller and fabricated in standardised modules in a factory, then shipped and installed. If that manufacturing-led model works, the cost overruns that have haunted large nuclear projects shrink. As of 2026 it is still being proven, which is the whole risk of the layer. These are bets on commercialisation, not on earnings.

How this layer relates to the rest. This layer is the theme's option on the far end of the time horizon, and it is the only part that could change the industry's economics rather than just its demand. It depends on everything else: on the fuel layer to supply it, on the engineering layer to manufacture it, and on the operating layer having demonstrated that buyers will pay for firm clean power. Nothing else in the theme depends on it, which is precisely why the theme can hold it without the whole roster riding on the outcome.

Between a licensed design and a working power plant sits a great deal of specialised manufacturing: pressure vessels and reactor components, fuel fabrication, turbines, and the grid equipment that connects any generator to a customer. This layer sells into the existing fleet, into government programs, and into whatever new capacity gets built, which is why it is the least clock-dependent part of the theme. It is also the layer with the most diluted exposure, because the same engineering capability that makes these firms useful to nuclear makes them useful to a dozen other markets.

How this layer relates to the rest. This is the theme's connective tissue and its partial hedge. It supplies the operators with components and services, it is what the SMR developers would have to hire to build anything at scale, and its government and non-nuclear work means it is funded regardless of whether commercial SMRs arrive on schedule. It moves for different reasons than either end of the roster, which is the point of including it.

The spine of this theme is not geography or company size. It is time. Constellation and Vistra are selling electricity from reactors that are running right now, so their nuclear exposure shows up in this year's results. Cameco is selling fuel under contracts written years before the power reaches a customer, so its clock is slower but still commercial. Oklo and NuScale have no meaningful revenue at all, because the reactors their value depends on have not been built. BWX Technologies and GE Vernova are paid across all of it, partly by government programs that do not wait on commercial nuclear at all.

Read that way, the dependency chain is clear. Data-center demand for firm carbon-free power is the input, and it hits the operating fleet first, because a buyer who needs round-the-clock clean electricity has to contract with someone who already owns reactors. That demand is also the commercial argument the SMR developers are making, so it flows outward to them on a much longer delay. Fuel sits underneath both, since a reactor is only as investable as the supply behind it, and equipment sits across both, because someone has to manufacture the components either way.

The practical consequence is that the seven names do not move for one reason. A licensing setback for an SMR design is close to fatal for the developer layer and almost irrelevant to Constellation's contracted output. A slump in the uranium price hits Cameco hardest and leaves BWX Technologies' naval work untouched. A drop in wholesale power prices runs straight through Vistra and barely touches a pre-revenue developer that has no power to sell. Understanding which shock lands on which layer is more useful than any ranking of the seven, and it is the reason the theme is described as a structure rather than a list.

A membership test is only credible if it excludes things. These are the names people most often expect to find here, and the specific reason each one does not qualify.

• Uranium miners and developers beyond Cameco. The near-term producers, developers, and explorers that make up most of the uranium trade sit in Walnut's separate uranium theme, not here. The distinction is what drives the stock: those names are leveraged bets on the uranium price and on getting a deposit into production, whereas this theme is about reactors and the power they sell. One large producer is in the roster as the fuel link in the chain; a roster of five miners would be a commodity portfolio wearing a nuclear label.
• The earliest-stage micro-reactor developers. Several very early micro-reactor and next-generation concepts trade publicly and appear in nuclear stock lists, including names discussed in the companion roundup. They are excluded here because the theme already carries two developers further along the path, and stacking a third pre-revenue design would tilt the roster toward the least-proven end rather than spanning the chain.
• Large regulated multi-utilities that own reactors. Several big regulated utilities, Duke Energy and Southern Company among them, own reactors as part of a much larger regulated fleet. They are excluded because the stock is driven by rate cases, allowed returns, and interest rates rather than by nuclear economics. The reactors are real; the exposure a shareholder gets to them is not the thesis.
• Hyperscalers signing nuclear power agreements. The technology companies contracting for firm clean power and, in some cases, backing reactor developers are the demand behind this theme, not members of it. Their revenue comes from cloud services and advertising, so a nuclear power agreement is a line in their cost base. They belong to cloud and AI themes, where that exposure is the thesis.
• Diversified engineering and construction contractors. Large EPC firms build nuclear projects, but nuclear is one contract type inside a backlog spanning refineries, chemical plants, and infrastructure. The inclusion test asks for material exposure, and a contractor whose results turn on non-nuclear project margins does not clear it, or the theme would absorb most of the industrials sector.

The uranium boundary is the one worth dwelling on, because it is where two Walnut themes touch. This theme is about reactors and the electricity they sell, and it holds exactly one fuel name, Cameco, as the link in the chain. Everything else on the fuel side, the near-term producers, the development-stage miners, and the explorers, sits in the uranium theme, where the driver is the commodity price and the execution risk of getting a deposit into production rather than the economics of generating power. If what interests you is the fuel market itself, that is the theme that expresses it, and the companion list is best uranium stocks. A company can be a good business and still be the wrong expression of a given theme.

The same seven names, grouped by the layer they occupy rather than ranked, so the shape of the theme is visible at a glance.

Three of the 7 earn revenue from nuclear today, two are pre-revenue developers, and two carry nuclear inside larger engineering businesses. That balance is the theme's central design decision, not an accident of what happened to be listed.

The passive route is a thematic fund. The theme names NLR (VanEck Uranium and Nuclear) and NUKZ (Range Nuclear Renaissance) as its proxies: NLR blends the utilities that run reactors with the uranium supply chain, while NUKZ reaches further into the reactor developers and the engineering names. Neither is a pure SMR play, because the SMR developers are small and pre-revenue, so an index that weights by size necessarily leaves them as a thin slice of the fund. Definitions vary widely across the category too: some funds lean heavily toward uranium miners, some toward utilities, and two funds with nuclear in the name can behave quite differently.

A theme inverts the trade. You know exactly which seven names you own, which layer each one represents, and what weight each carries, and you accept that seven names is a narrower roster than a fund holds. Neither is automatically better. The fund is the simpler instrument, the theme is the more deliberate one, and plenty of people hold a broad fund as a core with a small thematic tilt beside it.

A list of seven names is an input, not a portfolio. What turns one into the other is structure: which layers you want exposure to, what weight each name carries, and whether the concentration you end up with was chosen or inherited.

• Decide the layer mix first, then the names. The split between cash-generating operators and pre-revenue developers changes the character of the position far more than swapping one SMR developer for another.
• Set target weights that sum to 100. Equal weighting across seven names is a choice, and so is tilting toward the operators. Both are defensible. Not deciding is what leaves you concentrated by accident after one name runs.
• Size the speculative layer on purpose. The developer layer is the part with no revenue behind it, so whatever weight it carries should be a number you picked while calm rather than one that grew after a headline.
• Frame it against the S&P 500. A narrow thematic position should be judged against a broad benchmark, because the extra concentration has to be buying you something.
• Revisit as weights move. Thematic positions drift fast when the constituents have this much dispersion between them, and nuclear names have a habit of moving together on sentiment and then separating sharply on company news.

This is what Walnut is built for. You describe the thesis, the AI assistant proposes constituents and weights you can edit, the portfolio tracks as one performance line against the S&P 500, and you place trades you approve yourself at your own broker. Walnut is informational and does not tell you which stocks to buy.

For the companion view of which nuclear names are most widely held and discussed, see best nuclear stocks. For the fuel market on its own terms, see best uranium stocks.

The nuclear and SMR theme is seven companies across four layers, and the layering is the whole idea. Constellation Energy and Vistra run reactors that earn today. Cameco supplies the fuel every reactor consumes, contracted years ahead of the power. Oklo and NuScale Power are pre-revenue bets on small modular reactors becoming buildable at a cost the industry has never achieved. BWX Technologies and GE Vernova manufacture the components, fuel, turbines, and grid equipment that all of it depends on, funded partly by government work that does not wait on the commercial story.

Understood as a flat list of seven nuclear stocks, the theme looks like one concentrated bet on a single narrative. Understood as four layers running on different clocks, with different funding sources and different failure modes, it is a structure, and the structure is what you are deciding whether to own. Nothing here is a recommendation, and Walnut is not an investment adviser.