Fusion Energy Just Got a Lot More Real. Here's the 2026 Progress Report

Fusion power has been "thirty years away" for so long it became a running joke in energy circles. That joke is starting to look dated. In Devens, Massachusetts, a private fusion reactor called SPARC is now roughly three-quarters built — a real machine, mid-assembly, not a rendering in a pitch deck.

What SPARC actually is

SPARC is a compact tokamak built by Commonwealth Fusion Systems, a company spun out of MIT. Its defining bet is high-temperature superconductor magnets, which allow a much smaller and, in theory, cheaper reactor than older tokamak designs like the international ITER project — the core wager being that a smaller machine can prove the same physics without the multi-decade, multi-national construction timeline that's defined fusion research for generations.

The construction milestone

As of April 2026, Commonwealth Fusion Systems describes the facility as "about three quarters of the work to build SPARC" complete, spanning engineering, procurement, manufacturing, and assembly. Both halves of the reactor's vacuum vessel are now installed in the tokamak hall, along with two of the eighteen toroidal field magnets positioned on temporary assembly stands, and supporting infrastructure — the cryogenic plant, the RF building for plasma heating, and power systems — is already operational. What's left is exactly the hard, unglamorous part: measuring, cleaning, wiring, and eventually welding the vessel halves together before the machine can attempt to hold plasma at all.

The target that matters

Commonwealth Fusion Systems is targeting net fusion energy — the reactor producing more energy than it consumes — in 2027. The company is careful to frame this as a target, not a guarantee, and it's worth being precise about what it would actually prove: a physics milestone at the reactor level, not a commercially operating power plant delivering electricity to a grid at a competitive price. Those are two different and separately hard problems.

What's left is exactly the hard, unglamorous part: measuring, cleaning, wiring, and eventually welding the vessel halves together before the machine can attempt to hold plasma at all.

Real money is now behind it

This isn't purely a research project anymore. Energy company Eni signed a $1 billion offtake agreement in 2025 to purchase power from ARC, Commonwealth Fusion's planned commercial follow-on plant in Virginia — a real commercial commitment tied to a specific future facility, not just research funding. That kind of agreement is a meaningfully different signal than another grant or venture round: a large energy company is willing to contract for fusion power years before that power plant exists.

How this fits alongside the rest of the grid story

Fusion is a longer-horizon bet than either of the other two big power stories reshaping the grid right now: the small modular reactors racing toward the early 2030s, and the record-setting battery storage buildout already solving reliability problems today. If SPARC hits its 2027 target and ARC eventually follows, fusion offers something neither of those technologies does on its own: enormous baseload power without the long-lived radioactive waste of fission. That "if," at this stage, is still doing a lot of work — but for the first time in a long time, it's an "if" attached to a machine that's actually three-quarters built.