Back in December 2022, a lab in California did something people had been chasing for seventy years. It focused 192 laser beams onto a millimeter-scale capsule of hydrogen fuel and, for the first time, got more energy out of the reaction than the beams delivered to the target. The headlines called it a fusion breakthrough, and they were right to.<\/p>\n
What most of that coverage skipped over sits at the wall socket. To fire the 2.05-megajoule pulse that hit the target, the lasers at Lawrence Livermore’s National Ignition Facility pulled roughly 300 megajoules from the grid. The fusion gave back about 3.15 megajoules, close to the energy of three sticks of dynamite. So the reaction beat the beam, but the building as a whole lost around a hundred times more power than it made, and it could only fire once a day.<\/p>\n
A company down the road just opened a factory built to erase that gap. On July 10, Inertia Enterprises cut the ribbon on a 50,000-square-foot headquarters<\/a> in Livermore, a few miles from the lab where the 2022 shot happened. Inside are two things nobody has built at this scale before: a production line meant to mass-manufacture fusion fuel pellets, and the first pieces of a laser the company calls Thunderwall.<\/p>\n The building at 7800 Las Positas Road used to belong to Southwire, an electrical-products maker. Inertia leased it late last year, gutted it over about six months, and reopened it this month. Chief technology officer Mike Dunne has said the site won out partly because it was one of the few buildings around with enough electrical power to develop the lasers, which is the kind of detail that tells you what you are dealing with.<\/p>\n Inside, the company has already stood up a fuel-target manufacturing lab and a high-precision metrology facility. Inertia describes the site as the world’s first fusion fuel target factory, paired with what it calls the world’s most advanced high-energy laser system. Both efforts started in late 2025.<\/p>\n The target factory is not a marketing flourish. Right now, fusion fuel pellets are hand-made one at a time, each built to unique specifications for a single lab experiment. A power plant firing ten times a second would burn through targets by the million, and you cannot hand-make a million of anything to laboratory tolerances. So before you build the plant, you have to figure out how to stamp the pellets out like any other manufactured part.<\/p>\n The 2022 result was real, but it needs one word of precision: it was target<\/em> gain. The lasers delivered 2.05 megajoules to the fuel, and the fuel released 3.15 megajoules of fusion energy, a ratio of about 1.5. That met the formal definition of ignition, and the U.S. Department of Energy announced it<\/a> as the first controlled fusion experiment in history to cross that line.<\/p>\n Target gain is not the same as running a power plant. The 300 megajoules the lasers pulled from the wall to make that single 2-megajoule pulse never showed up in the celebration, but it is the whole problem. Measured at the fence line, the facility lost roughly a hundred times more energy than the fusion produced.<\/p>\n The reason is old hardware. NIF’s lasers are flashlamp-pumped, built on 1980s technology, and they were designed to prove a point once a day, not to run a city. That inefficiency is exactly what Inertia is aiming at.<\/p>\n Where NIF delivers about 2 megajoules every few hours, Inertia’s plant laser is designed to deliver 10 megajoules, ten times a second. In terms of average power, Dunne told Engineering News-Record<\/a>, that works out to roughly a million times NIF. Average power, not peak, is the number that matters. Firing one enormous pulse gives you a headline; firing continuously is what keeps the lights on.<\/p>\n The design gets there by going the opposite direction from NIF. Instead of one monster laser, Thunderwall is meant to be 1,000 smaller beamlines, each built in a factory, packed into a shipping-container-sized unit, and swapped out like a replaceable part. The company says each beamline runs about 20 times more efficiently than NIF’s and takes up a tenth of the footprint, thanks to diode-pumped lasers rather than flashlamps.<\/p>\n The efficiency target is the tell. Inertia says its prototype emitter is built to put out about 18 times the energy it consumes, and production versions will need to push past 30 times. That is facility-level gain, measured the way a utility would measure it, not the target-only number from 2022.<\/p>\n Most of the fusion machines you read about are on the other branch of the family tree: magnets, not lasers. The building-sized tokamaks like Japan’s JT-60SA<\/a> pin their plasma inside a ring of superconducting coils. A British magnet rig came apart<\/a> after more than 5,000 pulses, and a school-bus-sized magnet was shipped over from Spain<\/a> to revive an American reactor. All of them confine plasma with magnetic fields. Inertia crushes it with light.<\/p>\n Dunne’s argument for why that matters is about risk. Other fusion companies, he says, are still working to prove their approach can ever reach net gain, which is an open scientific question. Inertia is building on the one approach that has already shown it, at NIF, so the hard part left is engineering and manufacturing rather than physics. It is a competitive framing, and rival companies chasing magnetic and other designs would push back on the word “only,” but the underlying claim is straightforward: nobody else has produced target gain yet.<\/p>\nThe factory in an old wire plant<\/h2>\n
The number that took the shine off 2022<\/h2>\n
Thunderwall and the ten-shots-a-second problem<\/h2>\n
Why Inertia thinks it has a shortcut<\/h2>\n