Nuclear Energy Startups: Microreactors, Small Reactors and Fusion in 2026

Nuclear energy is having a revival. After decades in which few new reactors were built in the West, electricity demand from AI data centres, government support for clean firm power and a new generation of startups have brought nuclear back into the spotlight. Venture investors who once avoided the sector are now backing companies building factory-made microreactors, small modular reactors and fusion machines. This guide ranks the private nuclear energy startups in our database by sourced valuation, explains the different technologies and covers the considerable risks that remain.

Every company in the table links to a full profile with its sourced valuation, funding history and investors. We include private fission and fusion companies. Some leading companies, such as Commonwealth Fusion Systems and Pacific Fusion, have disclosed large funding totals but not valuations. Public nuclear companies such as Oklo and NuScale are discussed for context. For a fusion-only view, see our guide to fusion energy companies.

The ranking

Fission versus fusion

Today's nuclear power plants use fission: splitting heavy atoms such as uranium to release energy. Fission is proven technology that has supplied about a tenth of the world's electricity for decades. Startups are redesigning fission reactors to be smaller, safer and cheaper to build. Fusion, which powers the sun, joins light atoms together and promises abundant energy with less long-lived radioactive waste, but no fusion device has yet produced commercial power. Fission startups face mainly engineering, cost and regulatory challenges; fusion companies still face fundamental scientific and engineering hurdles. Investors in each are making very different bets on timing.

Microreactors and small reactors: Valar Atomics and Radiant

Valar Atomics is developing small high-temperature gas-cooled reactors for industry and data centres. Bloomberg reported it raised about US$450 million at a US$2 billion valuation in March 2026, and later reports described talks at a higher valuation after its test reactor achieved criticality in Utah. Radiant, founded by a former SpaceX engineer, is building Kaleidos, a portable one-megawatt microreactor designed to replace diesel generators at military bases and remote sites. It closed a Series D of more than US$300 million at more than US$1.8 billion in December 2025. Both reflect a shift toward factory production and speed.

Fusion leaders: Helion, Commonwealth Fusion Systems, Proxima and Pacific Fusion

Helion Energy, valued at US$15.5 billion after a June 2026 round led by Thrive Capital, is building a pulsed fusion system and has agreed to supply power to Microsoft, an ambitious commitment. Commonwealth Fusion Systems, an MIT spin-out, has raised billions of dollars to build its SPARC demonstration tokamak and plans a power plant in Virginia. Proxima Fusion, based in Munich, is designing stellarator power plants. Pacific Fusion raised a US$900 million Series A for a pulser-driven inertial fusion approach. Fusion companies have attracted the largest private funding totals in the energy sector, despite long timelines.

Why AI is reviving nuclear

AI data centres consume huge amounts of electricity around the clock. Technology companies have committed to clean energy goals, but solar and wind are intermittent, and battery storage cannot yet cover long periods economically. Nuclear power offers carbon-free electricity available at all hours. Major technology companies have signed agreements to restart old reactors, buy power from existing plants and support new reactor designs. These commitments give startups potential customers and credibility, although most new reactors will not be operating at scale until the 2030s.

Policy support

US policy has shifted strongly in favour of nuclear. Executive orders in 2025 aimed to speed up reactor approvals and expand nuclear capacity, and the Department of Energy launched a reactor pilot programme to get test reactors operating quickly, with a target of reaching criticality by July 2026. National laboratories provide testing facilities, and the military is exploring microreactors for bases. Other countries, including the UK, Canada, France and Japan, also support new nuclear development. Europe's fusion efforts benefit from strong public research programmes.

The fuel challenge

Many advanced reactors use high-assay low-enriched uranium (HALEU), which is enriched to higher levels than fuel for conventional reactors. Supply has been limited, and Russia was historically a major supplier, which became problematic after its invasion of Ukraine. The US government has funded efforts to build domestic HALEU production. Some designs also use TRISO fuel, coated particles that retain radioactive materials at high temperatures, which must be manufactured at scale. Fuel availability could become a bottleneck for deployment.

Regulation and licensing

Nuclear reactors must be licensed by regulators, such as the US Nuclear Regulatory Commission, which historically took many years to approve new designs. Reforms aim to shorten timelines, and test reactors at national laboratories can operate under Department of Energy authority. Commercial deployment still requires detailed safety cases, site approvals and public consultation. Companies that navigate licensing efficiently will have a major advantage.

How to evaluate a nuclear startup

Useful questions include: what technology does it use, and how mature is it? What has it demonstrated physically, such as criticality, plasma performance or component tests? What is its licensing path and timeline? How will it secure fuel? Who are its customers, and are agreements binding? How much capital will it need before generating revenue? And what will its power cost compared with alternatives?

Risks

Nuclear startups face technical risk, long timelines, regulatory uncertainty, fuel supply constraints and high capital requirements. Cost overruns have plagued large nuclear projects, and small reactors must prove they can be cheaper through factory production. Public acceptance varies by community. Fusion companies face the additional risk that commercial fusion may take much longer than expected. Valuations based on milestones rather than revenue can fall if progress slows.

What to watch

Watch test reactor milestones at national laboratories, licensing decisions, fuel supply agreements, power purchase deals with technology companies, fusion demonstration results from SPARC and Helion, and new funding rounds. Our guides to climate tech startups and AI infrastructure companies provide related context.

Key takeaways

Nuclear energy startups are attracting record investment, driven by AI electricity demand and policy support. Fusion companies such as Helion and Commonwealth Fusion Systems have raised the most, while fission startups such as Valar Atomics and Radiant aim to deploy sooner with small reactors. Technical, regulatory, fuel and cost challenges remain significant, and timelines should be treated with caution.

Small modular reactors explained

Small modular reactors, often shortened to SMRs, are fission reactors that generate up to roughly 300 megawatts of electricity, compared with more than 1,000 megawatts for a typical large plant. The idea is to build standardised modules in factories and assemble them on site, reducing construction time and cost overruns that have plagued large projects. Several designs exist. Some use water cooling like conventional reactors, while others use helium gas, molten salt or liquid metal, which can operate at higher temperatures and supply industrial heat as well as electricity. Microreactors, such as Radiant's Kaleidos, are an even smaller category, typically below about 20 megawatts, designed to be transportable.

The promise of SMRs depends on volume. A factory that builds dozens of identical reactors can improve quality and reduce costs through learning, but reaching that volume requires orders, and early units are usually expensive. Governments and large energy buyers can help by committing to multiple units. Several countries have announced programmes to deploy SMRs in the 2030s.

Who buys nuclear power from startups

Early customers for advanced reactors include the military, which values resilient power at bases; national laboratories and government agencies; remote mines and industrial sites that currently burn diesel; utilities seeking to replace retiring coal plants; and technology companies that need clean power for data centres. Each customer group has different requirements. Military customers prioritise security and resilience, industrial customers care about heat as well as electricity, and data centre developers want large, reliable supply at competitive prices. Startups often target one group first and expand later.

Lessons from nuclear history

The nuclear industry has a long history of ambitious promises and delayed projects. Large reactors in the US and Europe have suffered years of delays and billions of dollars in cost overruns, while some countries, including South Korea and China, have built reactors more predictably through standardised designs and experienced workforces. Startups say they have learned these lessons by designing for manufacturability and simplicity. Investors should still expect setbacks: first-of-a-kind projects almost always take longer and cost more than planned.

Kairos Power and Google

Kairos Power, which is building fluoride salt-cooled reactors at Oak Ridge, Tennessee, has an agreement to supply Google with up to 500 megawatts by 2035 and was reported in 2026 to have raised about US$643 million including Department of Energy funding. It has not disclosed a valuation. See our guide to data center power startups for how AI demand is reshaping energy investment.