MetroVolt's roadmap runs through G1: an integrated gate testbed costed at ~$0.5B (deposited breakdown: $582M ±30%) whose sole mission is to adjudicate the pre-registered predictions the design's value hangs on.
G1 exists to retire the named gaps in order of leverage: the confinement requirement (the H98 band), the hot-ion two-fluid balance (P6), and the exhaust/DEC physics chain. Seven predictions (P1–P7) are pre-registered in the series with acceptance criteria fixed before any experiment runs — the same discipline as the deposited Tier-2 decks, applied to hardware.
The cost breakdown is itself deposited (KX24 schedule): machine core, magnets at demonstrated field anchors, diagnostics, and facility, at gate scale rather than power-plant scale — which is what makes decisive physics affordable.
Staged capital against pre-registered milestones is how disciplined investors buy deep-tech risk. G1 converts 'is the physics right?' from an open-ended fear into a scoped, ~$0.5B, criteria-in-advance project — and every downstream billion waits on its verdict.
| G1 cost | ~$0.5B ($582M ±30%, deposited breakdown) |
| Mission | adjudicate P1–P7 pre-registered predictions |
| Highest-leverage gates | H98 band · P6 two-fluid · DEC chain |
| Criteria | fixed before first plasma |
| Capital discipline | FOAK waits on the verdict |