About Kronos

Two machines. The Hyperion breeder is a spherical tokamak built as a materials platform — its job is tritium and isotopes. The Aegis / MetroVolt burner is a D-³He tandem mirror, one core in two housings (defense installations and datacenter power). Together they lead with what a fusion machine makes on the way to electricity.

Kronos has a complete, frozen design, re-derived from first principles and published open access with its data and code. Construction begins in 2027; no machine has been built yet.

Net-electricity breakeven is the right long-term goal and the hardest one — a single, distant gate. Tritium and isotopes have real markets today, so leading with them turns one far-off milestone into a product that closes on tritium self-sufficiency. See the wedge.

The breeder (Hyperion) is a spherical tokamak whose gate is tritium self-sufficiency (breeding ratio up to 1.42 on-model) — it makes no net-electricity claim. The burner (Aegis/MetroVolt) is a low-neutron D-³He tandem mirror that closes on-model at engineering gain Q_E 1.318. Function first, always.

On the current plan: construction from 2027, first tritium around 2030-31, a WHAM-scale burner demonstration in 2032, and fleet build from 2036. Every date is a target tied to a gate, not a commitment — see the roadmap.

Yes. The De-Risking Register (164 gates), 141 interactive validators, the CGYRO dataset and a 44-paper series are all published with permanent DOIs. Independent reproduction is explicitly welcome — see reproducibility.

No. The physics comes from established, community-benchmarked codes. AI/ML sits alongside as method — surrogates, control, prediction — never as the source of the physics. A headline number is what a physics code computed, not what a model guessed. See how we compute.

Through the contact page. For the science, start at Science & Evidence.