Sweden's newest nuclear program has four reactors, five corporate partners, and one conspicuous absence: a final investment decision. On September 3, Studsvik AB signed an exclusive agreement with GE Vernova Hitachi Nuclear Energy and Samsung C&T to develop a four-unit BWRX-300 project totaling about 1.2 GW at one of its licensed Swedish sites. First power is penciled in for the mid-2030s.
The structure is the story. Studsvik, a Swedish nuclear technology company with roughly 250 employees, is not financing four reactors on its own balance sheet. It assembled a syndicate in which the reactor vendor, a construction giant, and two financial parties each carry a defined slice of the exposure.
It commits Studsvik, GE Vernova Hitachi Nuclear Energy, Samsung C&T, GE Vernova Financial Services and DS Investment Partners to advance a four-unit BWRX-300 program of about 1.2 GW in Sweden.
It is not a final investment decision, and it is not authorization to construct. The company confirmed the agreement is exclusive for a fixed period and extendable.
The syndicate model, not the reactor, is what the rest of Europe's small modular reactor pipeline now has to match.
Total capacity across four BWRX-300 units
Studsvik selected GE Vernova Hitachi and Samsung C&T after a competitive evaluation. · Studsvik AB, September 2026
The deal, stripped to its parts
The company leads the environmental impact assessment, site rights, community engagement, and the dialogue with the Swedish state. GE Vernova Hitachi owns reactor design, licensing support, and cost estimation. The remaining parties sit in design authority, advisory, and financial structuring. Each name maps to a task, and each task maps to a party that absorbs part of the cost of being wrong.
That division matters because a four-unit order is not four copies of one build. The consortium describes a phased, multi-unit structure meant to standardize work across the program and carry lessons from unit one into unit two. Standardization is the only lever that pulls the cost curve down before the fifth or sixth unit exists.
| Partner | Leads | Also carries |
|---|---|---|
| Studsvik | ✔ Site rights, impact assessment, state dialogue | ◐ Community engagement |
| GE Vernova Hitachi | ✔ Reactor design, licensing support, cost estimation | ◐ Supply chain |
| Samsung C&T | ✔ Engineering and construction delivery | ◐ Design authority |
| GE Vernova Financial Services, DS Investment Partners | ✔ Financial structuring and advisory | ◐ Design authority |
The ReFirm program did not appear from nowhere. It entered Studsvik through the acquisition of Kärnfull Next earlier in 2026. That developer had worked with GE Vernova Hitachi on BWRX-300 deployment in Sweden since 2022 and signed a teaming agreement with Samsung C&T in December 2024. The September announcement consolidates a relationship that has been assembling for four years.
"One reactor is a project. Four is the start of an industry."— Karl Thedéen, President and CEO, Studsvik AB
The policy window is real
Sweden passed legislation in 2026 that allows reactors beyond its three existing plants. That is the condition that makes any of this legible. Without it, the site at Nyköping or Målma in Valdemarsvik would face an approval wall no consortium can finance around.
It also holds U.S. Nuclear Regulatory Commission approval supporting next-generation reactor development, secured in June 2026, and applied for state support for new nuclear earlier in the year. The pattern across those filings is consistent: prove the licensing path first, then ask capital to follow.
As we wrote in August, the competition in small nuclear has moved to financing and schedule execution: who can fund a factory floor and hold a date. This agreement tests that thesis directly.
Why exclusivity is the real currency here
Confirmation criteria: a binding engineering, procurement and construction (EPC) contract signed before the exclusivity window closes.
What the mid-2030s timeline hides
Four units, first power in the mid-2030s. Read that against the calendar. The deal was signed in 2026. Environmental assessment, site rights, licensing, and a final investment decision sit between now and steel in the ground. The gap is the texture of every large infrastructure program.
That is also why the financial partners matter more than the technology headline. A project that cannot clear a final investment decision is a memorandum, not an asset. The presence of GE Vernova Financial Services and DS Investment Partners signals that the consortium expects to structure the build around project finance rather than a single corporate guarantee.
The comparison most readers will reach for is the BWRX-300 under construction at Ontario Power Generation's Darlington site in Canada, the first of its kind. That unit is the de facto reference plant. How it performs on cost and schedule sets the ceiling for what Sweden can promise.
What has to go right for first power in the mid-2030s?
Probability: 45% — the licensing groundwork and the existing partner relationships are in place, but no EPC contract or state risk-sharing package exists yet.
✅ Arguments for
The partner set has worked together since 2022, so the commercial relationships are not starting cold.
Sweden's 2026 legislation gives the program a legal basis that did not exist two years ago.
Confirmation criteria: a state support decision and a signed EPC contract.
❌ Arguments against
A reference plant in another country is not yet a proven cost curve.
The first unit's economics will be the worst of the four, and someone has to absorb that.
Disconfirmation criteria: exclusivity lapses without an EPC contract, or the state declines risk-sharing support.
Signals to track
A final investment decision, and whether it lands before the exclusivity period ends.
Sweden's decision on state support for new nuclear capacity.
Construction milestones at the Darlington BWRX-300 reference unit in Canada.
Conversion of the exclusivity agreement into a binding engineering, procurement and construction contract.
Development scenarios
🟢 Optimistic scenario (30%)
Implications: Sweden becomes the reference market for European BWRX-300 deployment, and the serial-build economics start to look real.
🟡 Base-case scenario (45%)
Implications: The program survives but loses its first-mover framing to faster-moving markets, and financing costs rise with the delay.
🔴 Pessimistic scenario (25%)
Implications: Studsvik returns to a single-site licensing story, and the Swedish SMR pipeline resets to zero committed capacity.