Spanish artificial intelligence startup Multiverse Computing announced a $570 million (€500 million) Series C funding round on July 27, 2026, reaching a $1.7 billion pre-money valuation and becoming one of Europe's newest AI unicorns. The round represents a fivefold step-up from the company's Series B and signals growing investor appetite for technologies that make AI dramatically cheaper to run.根据最新的 AI 行业报道，这笔资金是欧洲今年最大的 AI 轮融资之一。

受量子启发的模型压缩方法

At the heart of Multiverse's platform is CompactifAI, a compression technology that applies tensor networks — a mathematical framework originating from quantum physics — to the problem of AI model compression. The technique was pioneered by co-founder and Chief Scientific Officer Dr. Román Orús and can reduce the size of large language models by 80 to 95 percent with what the company describes as immaterial accuracy loss.

The result is AI that runs faster, consumes less energy, and can be deployed in environments where sending data to a hyperscale cloud provider is too expensive, too slow, or simply not permitted. Multiverse positions its technology as the missing efficiency layer that lets enterprises run powerful models on their own infrastructure rather than routing every workload through a major cloud platform.

"The AI industry has accepted a false constraint for years — that powerful models require expensive infrastructure," said Enrique Lizaso, co-founder and CEO of Multiverse Computing. "That constraint is gone. We have proven that AI can run at full performance on a smartphone, inside a sovereign data center, on a factory floor with no cloud connection."

重量级投资者阵容

C 轮融资由 Forgepoint Capital International、BNPP Solar Impulse Venture Fund 和 Bullhound Capital 共同领投。 Additional commitments came from a broad consortium including Santander Alternative Investments, Tikehau Capital, HP Inc., Orange Ventures, Scania Invest, NAventures (the corporate venture arm of the National Bank of Canada), Qatar Development Bank, Zouk Capital, SETT, EIC Fund, the Basque Government's Hazten Scale-Up Fund, and Kutxa Fundazioa.此轮融资由摩根大通和桑坦德 CIB 提供咨询。

包括之前的几轮融资在内，总资金预计将达到约 8 亿美元。该公司表示，此轮融资可能仍将开放以选择更多战略投资者。

Bullhound Capital 的联合创始人兼执行合伙人 Per Roman 以明显的欧洲术语来设计此次投资。 “多年来，欧洲一直担心自己无法在人工智能前沿竞争，”他说。 "Multiverse is the answer — a team out of San Sebastián whose technology lets anyone run powerful models on their own terms, on their own soil, without depending on anyone else."

收入爆炸性增长和实际部署

Since closing its Series B in June 2025, Multiverse Computing has grown annualized revenue by more than 10x, with 96x year-over-year sales growth in the first quarter of 2026 alone.这一发展轨迹使该公司跻身欧洲增长最快的人工智能基础设施企业之列。

Multiverse's compressed models are already deployed across millions of devices and systems, including drones, cameras, satellites, vehicles, and telecom infrastructure.该公司还致力于将其技术嵌入人工智能驱动的个人电脑中。 Its customer base of more than 100 organizations spans manufacturing, finance, energy, aerospace, cybersecurity, defense, and health and life sciences — including Allianz, Bank of Canada, Bosch, Iberdrola, Indra, PwC, and Telefónica.

Just days before the funding announcement, Multiverse disclosed that all CompactifAI-compressed models now run on Intel Xeon 6 processors, delivering what the company described as significant performance improvements, energy savings, and memory footprint reductions while preserving accuracy.

CompactifAI 平台

Multiverse 的平台涵盖高效人工智能的完整部署范围。压缩模型针对云、本地和设备上环境进行了优化，由 CompactifAI 路由器精心编排，实时决定给定的工作负载是应该在本地运行还是发送到云端。该公司还正在为下一代人工智能工厂和数据中心构建端到端软件层，将模型压缩、GPU 编排、人工智能服务和主权级控制集成到一个统一平台中。

对于企业来说，宣传很简单：Multiverse 不是更换现有基础设施或通过超大规模路由每个工作负载，而是在其之上添加一个效率层，将每个工作负载与正确的模型和后端相匹配 - 该公司声称可以节省数亿美元。

Forgepoint Capital International 董事总经理兼合伙人 Damien Henault 称 Multiverse 是“我们所见过的唯一一家既具备技术基础又具有商业吸引力的公司”，可以作为关键的 AI 基础设施平台，并指出它从下游 LLM 压缩工具演变为他所描述的完整的 AI 铸造厂和操作系统。

资金将支持什么

C 系列收益将用于资助 Multiverse 高效模型库的扩展、对专有压缩算法的持续研究、对主权人工智能超级工厂基础设施的战略投资，以及在东亚、东南亚、中东、加拿大和美国等关键市场建立更强大的区域业务。

该公司总部位于西班牙多诺斯蒂亚（圣塞巴斯蒂安），在美国、加拿大和欧洲各地设有办事处。

为什么人工智能效率现在很重要

Multiverse 的加薪是在人工智能行业面临计算成本螺旋上升之际进行的。训练和运行前沿模型需要巨大的 GPU 集群，企业越来越多地寻找在不产生超大规模费用的情况下部署人工智能的方法。模型压缩——减少大型模型的计算足迹，同时保留其能力——已成为该领域最具商业意义的研究方向之一。

BNPP Solar Impulse Venture Fund 管理合伙人 Yann Lagalaye 强调了可持续发展的角度。他表示：“除了为客户节省大量成本之外，Multiverse 的解决方案还通过先进的模型压缩技术，结构性地降低了 LLM 工作负载的能源强度。”他指出，GPU 利用率较低，能耗较低，基础设施要求也较低。

随着人工智能行业的成熟超越了原始能力竞赛，能够以较低成本提供同等性能的公司可能会成为最持久的赢家。凭借近 8 亿美元的总资金和不断增长的企业客户名册，MultiverseComputing 相信效率（而不仅仅是原始功率）将定义人工智能的下一阶段。

保持人工智能领先 阅读更多人工智能新闻 →