Multiverse Computing aims for $570m Series C at $1.7bn valuation to reduce AI costs
A Spanish deeptech company that crunches large language models wants investors to bet that success, rather than just measurement, is where some of the AI money is made.
Multiverse Computing, based in San Sebastián in Spain’s Basque Country, has opened a Series C round targeting up to $570m (€500m), the company said on July 27.
The raise will cost the startup around $1.7bn (€1.5bn) before new capital comes in, putting the company at around five times its previous valuation.
The superiority depends on CompactifAI, a tool that Multiverse says can compress a large language model by up to 95% with what it calls an imperceptible loss of accuracy. It borrows tensor networks from quantum physics to abstract the model, cutting down on memory, cost, and power each query fires.
💜 for EU tech
The latest talk from the EU tech scene, a story from our genius founder Boris, and some incredible AI art. It’s free, every week, in your inbox. Register now!
The strategy sits within a broader squeeze race that has attracted the biggest players.
Essentially, the company sells scaled-down versions of open models like Meta’s Llama, packaged to run on cheap hardware or locally instead of inside a hyperscaler data center.
That frame-to-cloud deployment is at the heart of the Series C pitch. As a consideration, the cost of actually running the model, surpasses training as the leading cost for most consumers, the argument goes, and they’ll pay for the same output with a fraction of the computing power.
It is not the only use of the Multiverse that has been found in mathematics. The company, which started with quantum software, has applied similar methods to other problems, including flood forecasting. A pass line compresses more to less, whether the target is a neural network or a river.
The round is led by Forgepoint Capital International, BNPP Solar Impulse Venture Fund, and Bullhound Capital, according to the company. If it closes at the top of the range, Multiverse’s total funding will reach around $800m across the rounds.
The company has yet to say when it expects to complete the raise, or disclose terms associated with the stakes of the leading investors. A declared open round can also close below its target.
In June 2025 the company closed a Series B worth 189 million euros, or about $215m, led by Bullhound with support from HP Tech Ventures, Toshiba, Forgepoint, SETT, and Spain’s CDP Venture Capital. Bloomberg reported in February that Multiverse was in talks for €1.5bn, the same amount now linked to Series C.
Multiverse was founded in 2019 by CEO Enrique Lizaso, a former banker, and Chief Scientific Officer Román Orús, a physicist whose work on tensor networks supports the product. Both the quantum business and its quantum tools reside within the European quantum push which has attracted significant public funding.
Its client list, by company, runs to Iberdrola, Bosch, Telefónica, Allianz, Bank of Canada, Indra, and PwC.
Multiverse also reported strong growth to match its ambitions, claiming a tenfold increase in annual revenue from the last round and first quarter sales up 96 times year over year. Those statistics are self-reported and have not been independently audited.
Betting is uniquely European. As the models grow more expensive to implement, a cluster of regional startups are chasing the bill of guesswork instead of the frontier.
The power demand of the data center has become one of the most difficult constraints in the industry, and reducing the energy consumed by the model per query is one of the few ways that startups can pull without having chips or power stations.
Multiverse sells reverse trading. If the future of AI is small, cheap, and close to the edge, a startup on the Basque coast would rather own that corner than chase the frontier quietly trying to shrink it down to size.
