SAN SEBASTIÁN, Spain, Dec. 02, 2025 (GLOBE NEWSWIRE) — As large language models continue to get larger and the scarcity of compute resources drives up development costs, AI is becoming increasingly cost-prohibitive for many enterprises. Today, Multiverse Computing, the leader in quantum-inspired AI model compression, and Cerebrium, an elastic, serverless AI infrastructure platform, announced a strategic partnership designed to alleviate this burden and create a new foundation for economically sustainable AI. Together, the technologies form a unified system that optimizes GPU utilization, minimizes latency, and lowers cost per inference across the full AI deployment lifecycle, from prototyping to production.
“Compute costs are still one of the biggest things holding back AI progress and limiting its impact on society,” said Enrique Lizaso, cofounder and CEO of Multiverse Computing. “We’re combining efficiency and scale so that innovation can finally take off without hitting a cost wall.” Our alliance shows that performance and cost can go hand in hand. This opens up AI that is faster, easier to get to, and more sustainable for everyone.
The main feature of the relationship is a smooth connection between Multiverse’s CompactifAI model compression engine and Cerebrium’s dynamic container scalability, which can quickly add thousands of GPUs. The collaborative solution lets businesses use high-performance models that run up to 12 times quicker, use up to 80% less computing power, and expand internationally in seconds, all without losing accuracy or availability.
Michael Louis, founder and CEO of Cerebrium, remarked, “Companies all over the world want to use AI, but very few can afford to do it on a large scale.” “As Multiverse’s compression engine reduces the amount of computing power needed and our infrastructure grows to meet demand, we’re getting rid of the last technical and economic barriers between innovation and real-world deployment.”
The alliance is also a sign of a bigger change in how the AI industry measures success. Instead of only looking at the number of parameters and the size of the models, it now looks at how well they work in terms of speed, efficiency, and cost-effectiveness.
Customers can now use Cerebrium’s elastic orchestration engine and CompactifAI models in private deployments.
More about Multiverse Computing
Multiverse Computing is the best at compressing AI models that are inspired by quantum mechanics. CompactifAI, a groundbreaking AI model compression engine, was created by the business because of its extensive knowledge of quantum software and AI. CompactifAI can shrink LLMs by up to 95% with only a 2–3% loss in accuracy. CompactifAI models lower the amount of computer power needed and open up new ways for AI to be used in many fields.
Donostia, Spain is home to Multiverse Computing’s main office, however the company also has operations in the U.S., Canada, and Europe. Multiverse Computing has raised about $250 million from investors like Bullhound Capital, HP Tech Ventures, SETT, Forgepoint Capital International, CDP Venture Capital, Toshiba, and Santander Climate VC. The company has over 160 patents and 100 customers around the world, including Iberdrola, Bosch, and the Bank of Canada.
About Cerebrium
Cerebrium is a serverless infrastructure platform that helps engineering teams quickly build and grow AI apps. The platform has fast startup times, can deploy to many regions for low latency and data residency, supports more than a dozen types of GPUs, and can scale to thousands of containers in seconds. Gradient (Google’s AI fund) and Y Combinator support Cerebrium, which runs real-time AI workloads for top teams including Tavus, Deepgram, and Vapi.
