Top Small Modular Reactor Stocks 2026: Factory-Built Nuclear

Traditional nuclear plants are hand-built cathedrals; SMRs are cars rolling off an assembly line.

Every SMR is under ~300 MWe, built from factory-made modules, shipped to wherever the power is needed: a data center campus, an industrial park, a retired coal site. Build the 10th unit cheaper than the 1st, the 100th cheaper than the 10th. That's the whole thesis.^[1]

In August, Oklo's Groves test reactor in Texas hit first criticality (a controlled chain reaction) just 11 months after breaking ground on an empty field. The first commercial-grade BWRX-300 is physically under construction at Darlington, Ontario. The DOE's Reactor Pilot Program is fast-tracking test reactors under federal (not NRC) authorization. Concrete is being poured. Fuel contracts are being signed.

Why now? Hyperscalers need gigawatts of firm, 24/7, carbon-free power, and they're writing checks: Google to Kairos, Amazon into X-energy, Meta to TerraPower. Washington wants US nuclear capacity quadrupled by 2050. The money, the policy, and the demand all showed up at the right time.

Some of these companies will be the next generation of utilities' favorite vendors. Some will run out of cash before their first reactor turns on.

Here's how we're slicing the theme: Pure-plays (reactor developers living or dying by their design), Integrated (giants where SMRs are one bet among many), Fuel & Components (picks-and-shovels that get paid no matter whose reactor wins), and the Private Bellwethers you can't buy yet, but should watch anyway.

These companies are the SMR trade. One design, one shot. If the reactor works and customers pay, the upside is enormous.

OkloNYSE: OKLO

If you've heard one SMR story this year, it's this one. In early August, Oklo's Groves isotope test reactor in Lockhart, Texas went critical just 11 months after groundbreaking. In an industry where projects run a decade late, that's the equivalent of a marathoner suddenly running a 4-minute mile.

The main event is the Aurora — a 75 MWe sodium-cooled fast reactor that can run on fresh HALEU or recycled nuclear waste. The first one is under construction at Idaho National Laboratory with DOE safety approvals stacking up (Nuclear Safety Design Agreement in March, Preliminary Documented Safety Analysis in June), targeting deployment by 2028. Oklo holds three slots in the DOE's Reactor Pilot Program — more than any other company — and just booked its first-ever revenue (~$1.2M in services). The catch: at a ~$8B market cap, you're paying up front for power sales that don't start until 2028 at the earliest. Execution is everything.

NuScale PowerNYSE: SMR

NuScale grabbed the ticker and the first-mover crown: it's still the only company with a US NRC-certified SMR design, the 77 MWe VOYGR light-water module. Boring, proven physics — which utilities love.

What NuScale doesn't have is a signed, funded US order. Its partner ENTRA1 is negotiating with the Tennessee Valley Authority on a power purchase agreement that could anchor up to 6 GW of deployments — potentially the largest nuclear program in US history — but as of Q2 2026, no definitive deal. The Romanian RoPower project (six modules at a former coal site in Doicești) cleared a shareholder vote but is still working through conditions, with final investment decision slipping to late 2026/early 2027. With ~$1B in liquidity, NuScale has runway. The stock is down sharply this year and short-sellers are circling; this is the "certified design, unproven business" trade. One signed TVA PPA changes everything.

X-energyNASDAQ: XE

The newest public pure-play — and it arrived with a bang. X-energy's April 2026 IPO priced at $23, well above the $16–19 range, raising over $1B alongside a private placement. Amazon's Climate Pledge Fund anchored the pre-IPO round, and the first Xe-100 units are slated for Dow's Seadrift, Texas site and near Washington's Columbia Generating Station.

Two things separate X-energy from the pack. First, real industrial customers: an 11.5 GW pipeline spanning Dow, Amazon, and the UK's Centrica. Second, it owns its fuel: subsidiary TRISO-X received a 40-year NRC special nuclear material license in February — the first new fuel fabrication facility licensed in over 50 years — and in August signed a definitive HALEU enrichment deal with Centrus. Reactor company + fuel company in one ticker.

Nano Nuclear EnergyNASDAQ: NNE

The lottery ticket of the group. Nano Nuclear is developing the Kronos MMR, a microreactor aimed at remote sites, military bases, and anywhere the grid doesn't reach — plus HALEU transport equipment and a fuel-supply MOU with Quadrant Nuclear Industries for a planned Idaho facility.

It's among the earliest-stage names on this list: no licensed design, no construction, years from revenue. Microreactors could be a huge market (they're small enough to truck to a customer), but NNE is a bet on management out-executing better-funded rivals. Size any position accordingly.

Hadron EnergyNASDAQ: HDRN

The smallest name here, in every sense. Hadron closed its SPAC merger with GigCapital7 in May 2026 and started trading as the first publicly traded light-water micro-modular reactor company. Its Halo MMR is a 10 MWe, factory-built, transportable reactor with a 10-year refueling cycle — think "power plant that arrives on a truck and runs for a decade."

The bull case: light-water physics (the same chemistry regulators have licensed for 70 years) shrunk to microreactor size, which could mean a smoother licensing path than exotic coolants. The bear case is right there in the deal terms: the merger raised roughly $31M, leaving about $24.5M in cash — enough for about a year of operations by the company's own math. That means dilution is a "when," not an "if," long before any Halo unit generates a watt. Even more speculative than NNE. Treat it as a tracking position on the micro-modular thesis, not a core holding.

Big companies where SMRs are a growth kicker. You give up torque; you get survivability.

GE VernovaNYSE: GEV

Want the SMR that's furthest along on Earth? It belongs to GE Vernova. The BWRX-300 — a 300 MWe boiling-water design from its GE Vernova Hitachi joint venture — is in full-scale construction at Ontario Power Generation's Darlington site, the first of four units, with the first targeted for completion by the end of 2030. In the US, the NRC is reviewing TVA's construction permit for a BWRX-300 at Clinch River, with a decision expected by December 2026.

The tradeoff is that GEV is a sprawling power-equipment giant (gas turbines, grid gear, wind), so the SMR business moves the needle slowly. But you're buying the design most likely to be generating commercial SMR power first in the Western world — inside a company that won't need to raise cash to survive.

BWX TechnologiesNYSE: BWXT

BWXT has been quietly building nuclear reactors for decades — for the US Navy. Every submarine and carrier reactor runs through its shops, which means BWXT gets paid real money, today, from a customer that never cancels.

The SMR angle: BWXT manufactures TRISO fuel, processed the HALEU for the first Pilot Program microreactors, is one of DOE's contracted HALEU deconversion providers, and builds components for other developers' reactors. Its own microreactor programs ride government funding. Think of BWXT as the machine shop the whole industry lines up outside of — profitable now, with SMRs as upside.

Most next-gen designs need HALEU — uranium enriched to 5–20%, versus ~5% for conventional reactors. Almost nobody in the West makes it. That scarcity is a business model.

Centrus EnergyNYSE: LEU

Every fast reactor and most advanced designs on this list share one dependency: HALEU. And Centrus operates the only NRC-licensed HALEU production facility in America, in Piketon, Ohio.

The customer list writes itself: an August 2026 definitive enrichment contract with X-energy (with prepayments funding Centrus's capacity buildout), a June letter of intent with Oklo covering fuel for up to five Aurora powerhouses from 2029, and a seat among DOE's HALEU enrichment contractors in awards worth up to $2.7B across the group, plus a ~$900M DOE contract in its backlog. The risks: a rich valuation (the market has noticed the chokepoint too) and eventual competition from Urenco, Orano, and newcomer General Matter. But if SMRs scale, Centrus owns the toll booth.

You can't buy these yet. Watch them anyway — they're where the biggest checks are landing, and at least two are heading for a ticker.

TerraPower (Private)

Bill Gates' reactor company is building the 345 MWe Natrium — sodium-cooled, with a molten-salt battery that can surge output to 500 MWe — at a coal town in Kemmerer, Wyoming, targeting operation around 2030–2031. In January, Meta agreed to support up to eight Natrium plants (as much as 2.8 GW), and in August Hyundai E&C signed on as EPC contractor for up to eight reactors with completion and price guarantees. TerraPower has engaged Goldman Sachs as a financial advisor — read into that what you will.

Kairos Power (Private)

Kairos builds a fluoride-salt-cooled, high-temperature reactor and has the industry's marquee tech partnership: a Google deal targeting 500 MW of capacity by 2035, starting with the Hermes 2 plant in Oak Ridge, Tennessee — construction started in 2026, ~50 MW to the TVA grid around 2030, under the first US utility PPA for a Gen IV reactor. Its Hermes test-reactor construction permit was extended to April 2029, a reminder that even the well-funded slip. IPO watchers peg Kairos as the most likely to follow X-energy onto Nasdaq, possibly as early as Q4 2026.

Holtec (NASDAQ: HNUC, pending)

The grown-up of the group. Holtec filed with the SEC for an IPO in July 2026 — reportedly targeting a $10B+ valuation — and unlike everyone above, it's already profitable: ~$165M in Q1 2026 revenue from spent-fuel storage, decommissioning, and the restarted Palisades plant in Michigan (the first US nuclear plant ever brought back from decommissioning). IPO proceeds are earmarked for its SMR-300 program, being advanced at Palisades and in the UK with EDF. If you want SMR exposure with an actual income statement attached, this is the listing to watch.

Two failure modes matter most.

First, FOAK economics. "First-of-a-kind" costs have already killed one flagship — NuScale's original Utah project (UAMPS) died in 2023 when projected power prices ballooned. The entire SMR thesis is the learning curve. If Darlington, Aurora-INL, or Natrium blow their budgets badly, the "assembly line" story collapses back into "cathedral, but smaller." Watch for final investment decisions that keep slipping (RoPower has already slipped a year), and the delta between promised and actual construction costs at Darlington.

Second, the fuel bottleneck. No HALEU, no fast reactors — full stop. Centrus's current output is measured in hundreds of kilograms per year; a commercial fleet needs metric tons. If enrichment capacity scales slower than reactor construction, finished reactors sit idle and the whole timeline shifts right. The things to track: Centrus expansion milestones, DOE HALEU allocations, and whether General Matter and Urenco actually deliver.

Honorable mention: political whiplash. This boom is riding executive orders and DOE fast-tracking. Policy giveth; policy can taketh away.

The next 12–18 months are catalyst-dense. Circle these:

NRC decision on TVA's Clinch River permit (expected December 2026). The first US construction permit for a BWRX-300 — a green light puts GE Vernova's design on American soil.

DOE Reactor Pilot Program criticality milestones. Oklo, Aalo, Valar and others racing to demonstrate test reactors under DOE authorization. Every criticality event is proof the fast-track model works — and a headline that moves the pure-plays.

A NuScale/ENTRA1–TVA definitive PPA — or not. Up to 6 GW hangs on this. Signed, it's the biggest order in SMR history. Stalled through 2027, and NuScale's "certified but customer-less" problem gets existential.

The IPO wave. Holtec's listing, possible Kairos and TerraPower offerings. Each one gives retail a new way in — and gives the market fresh price discovery on what private SMR bets are really worth.

Aurora-INL construction progress toward a 2028 startup. Oklo built a test reactor in 11 months. Doing it again — bigger, under commercial-grade scrutiny — is the proof point the whole sector needs.

Directionally? For the first time, the demand, the policy, and the construction are all real. But the market is pricing 2030 success in 2026. Own the theme through a barbell — one or two pure-plays you actually believe in, balanced with picks-and-shovels that get paid either way — and let the catalysts, not the hype, tell you when to add.