Verda Raises $189M Series B, Becomes AI Cloud Unicorn

Helsinki-based Verda has officially entered the European unicorn club, securing a substantial $189 million Series B funding round. This significant capital infusion propels the company, which designs and operates its own full-stack GPU data centers and AI platform, to a valuation that cements its status as a major player in the competitive AI infrastructure sector. The round was led by Emergence Capital, with crucial participation from MUFG Innovation Partners, Supermicro, Varma Mutual Pension Insurance Company, Lifeline Ventures, ENDUR, 6 Degrees Capital, byFounders, and Tesi, Finland’s state investment company. Notable angel investors, including Ola Tørudbakken, Director of AI Systems at Meta, and Mark Saroufim, co-founder of GPU MODE, also contributed.

Verda's unique approach bypasses the traditional reseller model by owning its infrastructure end-to-end, from data center construction to the proprietary software stack. This vertical integration allows for optimized GPU utilization and inference performance, addressing the rapidly evolving demands of AI workloads. The company’s platform already supports production AI for organizations across over 50 countries, with notable clients like Aleph Alpha utilizing its R&D infrastructure and Magnific powering millions of daily media generation requests. Epsilon Health also relies on Verda for training its radiology AI models at native resolution.

The strategic involvement of Supermicro, a key manufacturer with direct ties to Nvidia's accelerator supply chain, is particularly noteworthy. This partnership extends beyond capital, signaling a deeper integration and potential supply chain advantages for Verda. The company's rapid revenue growth is also a strong indicator of market traction, reporting an annualized revenue run rate of $165 million in July, a significant jump from $100 million just a month prior. This metric underscores the increasing demand for specialized, high-performance AI compute.

Founded in 2020 by Belgian engineer Ruben Bryon, Verda (formerly DataCrunch) has overcome significant market skepticism regarding the capital intensity and European origins of such ventures. Bryon, who built the company’s first GPU server rack himself, was joined by co-founders Milosz Szewczak and Tamir Segal. This Series B follows a series of successful funding rounds, including a $13 million seed in 2024, a $64 million Series A in 2025, and a $117 million equity-and-debt package in April 2026, later expanded to $155 million with financing from the Nordic Investment Bank. The total capital raised now exceeds $450 million.

Verda plans to leverage this new funding to expand its operational capacity to over 250 megawatts by 2027. This expansion will target new data center locations across Europe, the UK, the US, and Asia. Key investments will also focus on early deployments of Nvidia’s VR200 NVL72 rack-scale systems, enhancing provisioning speeds, and bolstering enterprise-grade security features such as audit logs and single sign-on (SSO).

While Verda's growth is impressive, it operates in a market with established giants like Crusoe, which recently raised $3.9 billion at a $30.9 billion valuation, and CoreWeave, with its massive $21 billion infrastructure agreement with Meta. Independent GPU cloud providers like FluidStack and Nscale are also rapidly scaling. Verda's competitive edge lies in its strategic focus on European clients who prioritize data sovereignty and seek alternatives to US hyperscalers. The company's success will hinge on its ability to scale globally while maintaining competitive pricing and high equipment utilization, a challenge that will be closely watched by the broader AI infrastructure sector.