Skip to content
ContentLora

    Tip: press / anywhere to search.

    organization

    Commonwealth Fusion Systems

    Also known as CFS

    Commonwealth Fusion Systems (CFS) is a company based in Devens, Massachusetts, that builds compact tokamaks with high-temperature superconducting magnets. In 2021 it tested a 20-tesla magnet with MIT.[1][2][3] Its SPARC machine, which CFS said was almost 80 percent complete in August 2026, aims to show net fusion energy (Q>1), a result CFS publicly targets for 2027. CFS plans a 400 MW ARC power plant in Virginia for the early 2030s and says it has now raised $4 billion.[1][4][5][6]

    Editor reviewedUpdated Fusion energyEnergy and climatePhysics
    Key facts

    What CFS does

    Commonwealth Fusion Systems builds tokamak fusion machines around hts-fusion-magnets. These magnets are wound from REBCO tape, a high-temperature superconductor that CFS describes as having recently reached industrial maturity.[7] Between 2018 and 2021, CFS and MIT’s Plasma Science and Fusion Center developed the SPARC Toroidal Field Model Coil.[3] On 5 September 2021 the coil reached 20 tesla, which MIT called the most powerful magnetic field of its kind created on Earth.[2] MIT said such magnets allow the performance of a conventional device about 40 times larger in volume.[8]

    The 2021 test results, including a deliberately induced quench, were published in six peer-reviewed papers in IEEE Transactions on Applied Superconductivity in March 2024.[9] MIT’s Dennis Whyte said the result changed the cost per watt of a fusion reactor by a factor of almost 40.[10]

    SPARC

    SPARC is being built at CFS’s headquarters in Devens, Massachusetts. CFS says it will be the first commercially relevant machine to demonstrate net fusion energy, meaning Q>1.[11] On 5 August 2026 CFS said SPARC was almost 80 percent complete.[1] Its target is lower than iter‘s Q of at least 10. HTS magnets allow it to be much smaller, and CFS calls it commercially relevant.[12][11][13]

    CFS does not publish a dated first-plasma milestone. Its public schedule is stated as net energy: company pages said in October 2026 that in 2027 SPARC would become the first commercially relevant machine to get more energy out of fusion than is needed to drive it.[4] A company blog post current at the same date said operations would start “in coming months” and that because SPARC is designed for Q>10, Q>1 should not be an extremely difficult milestone.[14] Both statements come from CFS, and no independent schedule verification exists.

    CFS is also building a software model of the machine. On 6 January 2026 it said it would create a SPARC digital twin with nvidia and Siemens, combining Siemens engineering software with NVIDIA Omniverse libraries and OpenUSD.[15] On 8 October 2026 CFS said the US Department of Energy had chosen it as the only private company leading a project in the second phase of the Genesis Mission, for an AI-enabled digital twin platform to optimise SPARC operations.[16]

    ARC and customers

    In December 2024 CFS chose the James River Industrial Center in Chesterfield County, Virginia, for its first commercial plant, ARC. ARC is planned to generate about 400 megawatts and to deliver power to the grid in the early 2030s.[5] On 30 June 2025 Google agreed to buy 200 megawatts of ARC’s output and made a new investment in the company.[17] CFS reports that it has cleared two ARC milestones under DOE’s milestone programme: a preconceptual design and a technology roadmap.[18]

    Supply chain preparation has started. On 30 September 2026 CFS agreed to buy more than 10,000 kilometres of HTS tape from Fujikura for ARC plants, starting with the Fall Line Fusion Power Station in Virginia. The companies called it the largest agreement of its kind.[19]

    On 4 June 2026 CFS published five peer-reviewed physics basis papers for ARC. It said the analysis gives confidence that ARC can continuously deliver 400 megawatts of net electricity, while marking out the physics areas where SPARC data will reduce uncertainty.[20] Fuel supply is a separate problem: a deuterium-tritium plant must breed more tritium than it burns, because today’s civilian tritium stock is only tens of kilograms.[21][22] On 1 July 2026 the UK Atomic Energy Authority named CFS the first international company in LIBRTI, a 220-million-pound programme aiming to demonstrate net tritium production, with CFS building the first test articles.[23]

    Funding

    On 30 July 2026 CFS said a further $1 billion of equity brought its total capital raised to $4 billion, about 30 percent of all capital raised by the fusion industry on its own count.[6] It called the round the largest in fusion since its own $1.8 billion Series B in 2021.[24] The Fusion Industry Association separately lists CFS’s $863 million Series B2 round in August 2025 among the industry’s largest recent raises, and puts total private fusion funding at $14.24 billion by July 2026.[25][26]

    What to watch

    The main milestone to watch is whether SPARC reaches Q>1, which CFS targets for 2027.[4] CFS expects ARC to deliver power in the early 2030s.[5] In August 2026 CFS said SPARC was still under construction.[1] Competing schedules are worth watching alongside it: helion-energy says its Orion plant is designed for initial operations in 2028, while iter does not start deuterium-tritium operation until 2039.[27][28] The debate over these dates is set out in the fusion timeline debate.

    Questions readers ask

    What is SPARC?

    SPARC is CFS's compact high-field tokamak in Devens, Massachusetts. CFS intends it to be the first commercially relevant machine to show net fusion energy (Q>1). CFS said in August 2026 that it was almost 80 percent complete.[11][1]

    When does CFS plan to put power on the grid?

    CFS plans its first ARC power plant, about 400 megawatts, in Chesterfield County, Virginia, and expects it to deliver power to the grid in the early 2030s.[5]

    Who has agreed to buy CFS's power?

    Google signed a 200-megawatt power purchase agreement for ARC in June 2025, which CFS called the largest fusion power purchase agreement to date.[17]

    How much money has CFS raised?

    CFS said on 30 July 2026 that a further $1 billion of equity brought its total capital raised to $4 billion, which it called about 30 percent of all capital raised by the fusion industry. The Fusion Industry Association lists its earlier $863 million Series B2 round in August 2025.[6][25]

    When will SPARC produce its first plasma?

    CFS has not published a dated first-plasma milestone. Its public target is Q>1 in 2027, and a company blog post current in October 2026 said operations would start in coming months.[4][14]

    Sources

    Each numbered claim is a statement we checked against the sources listed with it. Status shows how well established it is.

    1. [1]

      As of 5 August 2026 CFS said SPARC was almost 80 percent complete. reportedas of 2026-08-05

    2. [2]

      On 5 September 2021 MIT and Commonwealth Fusion Systems ramped a large high-temperature superconducting magnet to 20 tesla, the strongest fusion magnet of its kind. confirmedas of 2026-10-10

    3. [3]

      The 20-tesla magnet tested in 2021 was the SPARC Toroidal Field Model Coil, built by MIT's Plasma Science and Fusion Center and CFS between 2018 and 2021 to develop REBCO magnet technology for the SPARC tokamak. confirmedas of 2026-10-10

    4. [4]

      As of October 2026 Commonwealth Fusion Systems publicly targets 2027 for SPARC to produce more energy from fusion than is needed to power the process, the Q>1 threshold. reportedas of 2026-10-10· forecast

    5. [5]

      In December 2024 CFS chose the James River Industrial Center in Chesterfield County, Virginia, for ARC, a planned power plant of about 400 megawatts expected to deliver power to the grid in the early 2030s. confirmedas of 2026-10-10

    6. [6]

      On 30 July 2026 CFS said it had raised a further $1 billion of equity, bringing its total capital raised to $4 billion, which it described as about 30 percent of all capital raised by the fusion industry to date. reportedas of 2026-07-30

    7. [7]

      Commonwealth Fusion Systems describes REBCO (rare-earth barium copper oxide) as a high-temperature superconductor that has recently reached industrial maturity. confirmedas of 2026-10-10

      • Technology · Commonwealth Fusion Systems (retrieved 2026-10-10)
    8. [8]

      MIT said high-temperature superconducting magnets allow the same performance as a device about 40 times larger in volume built with conventional low-temperature superconducting magnets. confirmedas of 2026-10-10

    9. [9]

      The MIT-CFS magnet removed the insulation normally wound between superconducting tape layers, and the 2021 test results were published in six peer-reviewed papers in IEEE Transactions on Applied Superconductivity in March 2024, including a deliberately induced quench. confirmedas of 2024-03-04

    10. [10]

      MIT's Dennis Whyte said the 2021 magnet result changed the cost per watt of a fusion reactor by a factor of almost 40. reportedas of 2024-03-04· interpretation

    11. [11]

      Commonwealth Fusion Systems says its SPARC tokamak in Devens, Massachusetts, is intended to be the first commercially relevant machine to demonstrate net fusion energy (Q>1). confirmedas of 2026-08-05

    12. [12]

      ITER's main performance goal, Q of at least 10, means 500 megawatts of thermal fusion power from 50 megawatts of heating power put into the plasma, in 400-second pulses. confirmedas of 2026-10-10

    13. [13]

      A 2024 analysis of the CFS magnet found REBCO superconductors well suited to tokamaks, enabling devices some 30 to 40 times smaller in volume than earlier designs. confirmedas of 2024-04-04

    14. [14]

      In a company blog post current as of October 2026, CFS said it would start operating SPARC "in coming months" and that SPARC is designed to reach Q>10, so Q>1 is not an extremely difficult milestone; CFS has not published a dated first-plasma target. reportedas of 2026-10-10

    15. [15]

      CFS is building a digital twin of SPARC with NVIDIA and Siemens, announced on 6 January 2026, using NVIDIA Omniverse libraries and OpenUSD together with Siemens engineering software. reportedas of 2026-01-06

    16. [16]

      On 8 October 2026 CFS said the US Department of Energy had selected it as the only private company leading a project in the second phase of the Genesis Mission, for an AI-enabled digital twin platform to optimise SPARC operations. reportedas of 2026-10-08

    17. [17]

      On 30 June 2025 Google agreed to buy 200 megawatts from CFS's ARC plant, which CFS called the largest fusion power purchase agreement to date, and made a new investment in the company. confirmedas of 2026-10-10

    18. [18]

      CFS reported clearing two ARC milestones in DOE's milestone programme, a preconceptual design and a technology roadmap. confirmedas of 2026-08-05

    19. [19]

      On 30 September 2026 CFS agreed to buy more than 10,000 kilometres of HTS tape from Fujikura for its ARC power plants, starting with the Fall Line Fusion Power Station in Chesterfield County, Virginia; the companies called it the largest agreement of its kind, and Fujikura is expanding capacity to supply it. confirmedas of 2026-10-10

    20. [20]

      On 4 June 2026 CFS published five peer-reviewed physics basis papers for ARC, which it said give confidence that the plant can continuously deliver 400 megawatts of net electricity while identifying physics areas where SPARC data will reduce uncertainty. reportedas of 2026-06-04

    21. [21]

      To be self-sufficient in fuel, deuterium-tritium power plants need a tritium breeding ratio greater than one, and lithium-6, the isotope most important for breeding, is only about 7.6% of natural lithium. confirmedas of 2024-05-20

    22. [22]

      Today's civilian tritium comes mainly from heavy-water fission reactors such as Canada's CANDU units, with an estimated 20 to 30 kilograms available. confirmedas of 2024-05-20

    23. [23]

      On 1 July 2026 the UK Atomic Energy Authority named CFS the first international company to join LIBRTI, a 220-million-pound programme aiming to demonstrate net tritium production, with CFS building the first test articles. confirmedas of 2026-07-01

    24. [24]

      CFS called its July 2026 round the single largest funding round among fusion energy companies worldwide since its $1.8 billion Series B in 2021. reportedas of 2026-07-30

    25. [25]

      Large recent rounds include CFS's $863 million Series B2 (August 2025), Proxima Fusion's $518 million (July 2026), Helion's $465 million (June 2026) and Inertia Enterprises' $450 million Series A (February 2026). confirmedas of 2026-07-13

    26. [26]

      In the 12 months to July 2026, 56 fusion companies raised $4.48 billion, bringing total private fusion funding reported since 2021 to $14.24 billion. confirmedas of 2026-07-13

    27. [27]

      Helion's first commercial plant, Orion, in Chelan County (Malaga), Washington, is designed to deliver at least 50 megawatts and begin initial operations in 2028. confirmedas of 2026-10-10

      • Orion · Helion Energy (retrieved 2026-10-10)
    28. [28]

      ITER's 2024 baseline schedules the Start of Research Operation in 2034, deuterium-deuterium operation in 2035, full magnetic energy in 2036 and the start of deuterium-tritium operation in 2039, about four years later than the 2016 plan. confirmedas of 2026-10-10

    Revision history (2)
    1. Page created.
    2. Refresh: added the 2027 Q>1 target, the $1bn July 2026 round and $4bn total, the five ARC physics papers, the LIBRTI tritium-breeding partnership, the NVIDIA and Siemens digital twin and the DOE Genesis Mission project.

    Created Oct 10, 2026. Last reviewed by an editor on Oct 10, 2026. Next scheduled review: Jan 10, 2027.

    Cite this page

    "Commonwealth Fusion Systems." ContentLora, updated Oct 10, 2026. https://contentlora.com/wiki/commonwealth-fusion-systems

    Spotted an error? Suggest a correction or emailcorrections@contentlora.com.