30 Sept 2026

Zenithon raises a £7.5m seed round led by BACKED for ML models that solves complex physics

Zenithon builds proprietary world models for extreme physics that learn from simulations and real-world experiments. Engineers use Zenithon to explore large design spaces, search beyond familiar designs and improve the performance of complex machines.

Zenithon, a startup building world models for extreme physics, has raised £7.5 million in a seed round led by BACKED, with participation from Lunar, Seraphim, MMC and SOSV, alongside founders and hyperscaler directors. Funding supports training large world models for extreme physics and expanding the team in San Francisco and the United States, as well as growing the research team in London.

Zenithon builds proprietary models that can explore a million design points in the time it takes to simulate one and learn directly from real-world experiments. Engineers use Zenithon to search beyond familiar designs, discover new levels of performance and keep improving how complex machines operate across fusion reactors, hypersonic applications and advanced semiconductor manufacturing. Technology spans the data generation layer through model architecture, training, inference and delivery, with new models planned every three months.

Alongside AI, this century is shaped by rockets, fusion reactors and fabs, where building the next generation requires understanding and controlling increasingly complex physics. Current tools struggle when a single simulation or experiment takes days, limiting even the best engineering teams to exploring only a fraction of the possibilities. Zenithon focuses on extreme physics problems such as fusion reactors, shocks for hypersonics and the most demanding semiconductor environments, aiming to keep human innovation on an exponential path as technologies become more complex.

Zenithon was founded formally in July 2025 and is co-founded by Alex Higginbottom and Abetharan Antony. The team brings together researchers behind foundational breakthroughs in machine learning for physics and a wider group from leading labs globally. Zenithon is currently a full-time team of 11, with plans to expand to 17 over the next 3 to 6 months and to grow research capacity in London while building a delivery team in San Francisco and the United States.

Use of funds is split roughly half on headcount and half on compute, with each new generation of models requiring significant compute and capital. Zenithon plans to continue working with the United Kingdom fusion, aerospace and advanced semiconductor ecosystems to develop strong sovereign capability in critical industries.

We are delighted to be partnering with Alex and Abi as they develop SOTA Foundation Models for extreme physics. Advancements in fusion, hypersonics and chip manufacturing - critical industries for the advancement of humanity - are held back by the cost, speed and accuracy of current simulation and design methods and we believe Zenithon have assembled the best team globally to make breakthroughs in this area. We have seen first-hand how AI is revolutionising complex manufacturing through investments in companies like Flow, Olix and CloudNC. We believe Zenithon will have a profound impact on these industries and are thrilled to be partnering with this exceptional team.

Alex Brunicki, Partner at BACKED
Powered by
NatWestSageHarmonicNovus

Similar articles