NIF reached a target gain of 4.1 and no company has one

One gain figure, and only one

A National Ignition Facility shot reached a record 8.6 MJ fusion yield at a target gain of 4.1 and a hot-spot temperature of 20.1 keV, up to 2.2 times the neutron yield of its closest comparison design, reported in Physical Review Letters. What changed was ramped-profile doping of the capsule.

It is also the only measured gain figure anywhere in this ledger's fusion record. Nine fusion companies are on file here. Not one of them has published a gain result of its own, and most do not claim to be measuring that quantity at all.

What the companies report instead

• **Helion Energy** ran its Polaris prototype at 150 million °C in January, the first privately developed machine to demonstrate measurable deuterium-tritium fusion, and its chief executive declined to give a scientific breakeven figure, saying the company measures itself on electricity rather than pure scientific milestones.
• **Commonwealth Fusion Systems** targets Q>1 in 2027, which is gain in the plasma and not electricity: SPARC has no generator, and the machine meant to sell power is ARC, a 1.1GW design converting to 400MW net to the grid.
• **Pacific Fusion**'s Sirius pulser has passed 3,000 shots at Lawrence Livermore, delivering 60GW into a resistive load in a 100 nanosecond pulse at 95% energy efficiency. Its fusion numbers are targets: bursts above 100 MJ, and net facility gain, a facility-level threshold, by 2030.
• **Focused Energy** raised a $240m Series A led by RWE for a laser fusion plant on the decommissioned nuclear site at Biblis by the mid-2030s. Like every laser fusion company outside the National Ignition Facility, it has published no ignition or gain result of its own.
• **First Light Fusion** stopped being a plant developer in March 2025 and now supplies the companies that still are, starting with velocity amplifiers sold to Texas A&M.

Temperature, tesla, shot counts, pulser efficiency, tritium breeding ratios: the private record is full of real measurements. None of them is a gain figure, and the companies are not pretending otherwise.

The money is dated, not measured

**Helion Energy** is valued at $15.5bn after a $465M Series G led by Thrive Capital, with $1.5bn raised in total, holds the first regulatory approvals anywhere to operate a fusion power plant, and has contracted a 50MW machine to supply Microsoft from 2028. **Commonwealth Fusion Systems** has raised $4bn, more than any other private fusion company. **Proxima Fusion** raised €411M at a €2.4bn valuation for a net-energy demonstrator in the early 2030s. **TAE Technologies** says its first commercial-scale plant will produce fusion power in 2031.

Six dates, running from 2027 to the mid-2030s, and three different thresholds behind them: gain in the plasma, gain across a facility, and electricity on a grid. They are not the same achievement, and clearing the first does not imply either of the others.

How to read today's number

The 4.1 belongs to a machine with no customer and no revenue date, and it was won by changing the dopant profile in a capsule rather than by building anything larger. That is the shape of the only progress the field can currently prove: incremental, published, and at a facility that is not trying to sell power.

The private dates are not discredited by that. They are simply dated against thresholds that nobody, including the facility holding the record, has reached. The useful question on each of the six is not whether the physics works but which threshold the valuation was priced against, and whether the company says which one it means. On today's record, Pacific Fusion and Commonwealth Fusion Systems name theirs precisely, and Helion Energy declines to give one at all while carrying the earliest delivery date in the file.