Building pulsed-power fusion power plants to deliver abundant, clean, firm energy
HQ: Fremont, CA | Founded: 2023 | Employees: ~158 (mid-2026, +106% YoY) | Stage: Early-to-mid stage VC (Series A, milestone-gated) | Website: https://pacificfusion.com
Pacific Fusion was quietly founded in the summer of 2023 by a small group of national-lab scientists and technologists who wrote and then set out to prove the ideas in a May 2023 fusion white paper from Science for America, the non-profit founded by Eric Lander. Lander — who led the US Human Genome Project, co-founded the Broad Institute, and served as President Biden's science adviser (OSTP director) — co-founded the company with Will Regan (ex-ARPA-E and Google X), Keith LeChien (former DOE/LLNL, co-inventor of Pacific Fusion's pulsed-power architecture with LLNL's Bill Stygar), Carrie von Muench (ex-BCG/Google X/Innovation Endeavors), and Leland Ellison (former LLNL simulation scientist).
The company's reactor concept, CHROMA, is a pulsed-magnetic inertial fusion device: micron-scale deuterium-tritium fuel is confined inside cylindrical metal liners that a massive electrical pulse — from 'impedance-matched Marx generator' (IMG) modules — magnetically implodes at ~100 km/s, compressing and heating the fuel to fusion conditions. This scales Sandia's MagLIF concept beyond the Z Machine's ~20 MA limit. In an April 2025 peer-reviewed paper (arXiv 2504.10680; Physics of Plasmas, Sept 2025), Pacific Fusion laid out a Demonstration System driver with >60 MA current and ~80 MJ stored energy, claiming ~1,000x the effective performance of NIF per unit, ~100x facility-level energy gain (Qf > 1), at roughly 1/10th NIF capital cost, using 156 identical IMG modules delivering ~2 terawatts for ~100 nanoseconds.
Pacific Fusion exited stealth on October 25, 2024 with a more-than-$900M Series A led by General Catalyst (Hemant Taneja), with Breakthrough Energy Ventures, Lightspeed, Lowercarbon Capital, and individuals including Eric Schmidt, John Doerr, Ken Griffin, Patrick Collison, Reid Hoffman, and Mustafa Suleyman. Structured like biotech financing, the round releases in tranches against pre-defined technical milestones: Tranche 1 unlocked in November 2024 (first-phase milestones completed ahead of schedule) and Tranche 2 in June 2026 after a one-third-scale prototype pulsed-power module delivered ~440 GW peak power at ~1.1 MV in an 80-ns burst — the 'highest-power single-step pulsed-power driver ever demonstrated' (11x the scale of LLNL's Sirius prototype). Under a CRADA, LLNL's Sirius IMG prototype surpassed 3,000 shots by July 2026, and in August 2026 Pacific Fusion broke ground on a $1B Research & Manufacturing campus at Albuquerque's Mesa del Sol (with ~$776.6M in New Mexico industrial revenue bonds) and signed an NNSA Memorandum of Understanding on high-yield fusion and pulsed-power science.
The company is building toward a full-scale >1 TW pulser module, then a 156-module Demonstration System targeting net facility gain (high-yield >100 MJ fusion bursts) by 2030, and expects its first commercial power plants in the mid-2030s across the US Southwest, France, and Bangladesh (team distributed in three countries). It competes in a private-fusion market that passed $11.5–13B in total funding (June 2026), distinguished by capital efficiency, milestone accountability, and national-security value.
Pacific Fusion is the most technically audacious bet in fusion: it claims a patent-protected pulsed-magnetic architecture that can hit ~100x facility-level energy gain at roughly one-tenth the capital cost of NIF-style inertial fusion — and it structured its more-than-$1B Series A so that capital releases only against technical milestones, a biotech-style model applied to hard energy. The team's credibility is exceptional (Eric Lander, genome-mapping scientist and former White House science adviser; Keith LeChien and Leland Ellison from LLNL; deep national-lab ties). The June 2026 440 GW prototype and the thousand-shot-plus LLNL 'Sirius' campaign are genuine, verifiable progress. But the road is long: the next gating milestone is a full-scale ~1 TW module, a full Demonstration System needs 156 modules, and no fusion-yield 'compression shot' or any offtake/PPA exists yet — making Pacific Fusion a high-upside, high-science-risk position whose Series-A valuation is undisclosed and whose payoff is a Decade-plus away.
Browse: Home | Blog | About | All Startups A-Z | View full profile on StartupWiki