Helsinki’s DataCrunch, a provider of high-performance AI infrastructure, has raised a total of €55 million in a Series A round to scale its compute platform and further its ambition to become Europe’s first AI cloud hyperscaler.
This round was led by byFounders, Skaala, Varma pension fund, and Tesi, with additional participation from J12 Ventures and select angel investors. A portion of the round includes debt financing led by Nordea and Armada Credit Partners with participation from Danske Bank, Norion Bank, and Local Tapiola.
“European organisations are facing a critical choice: continue depending on foreign cloud providers or invest in local infrastructure that offers true data sovereignty,” said Ruben Bryon, CEO and Co-founder of DataCrunch.
Founded in 2020 by CEO Ruben Bryon, this Series A brings DataCrunch’s total capital raised to €76.5 million.
DataCrunch provides high-performance, cost-effective cloud GPU solutions for AI training and inference. Current customers include Sony, Freepik, Schibsted, 1X, Unbabel, and multiple educational institutions.
The company is building a one-stop platform for AI development, enabling fast, affordable access to compute clusters optimised for AI workloads; supporting everything from research and training to inference at scale. DataCrunch is among the first globally to deploy NVIDIA’s HGX B200 systems and will soon integrate the B300 and GB300.
Bryon added: “This funding enables us to accelerate our vision of becoming Europe’s first AI cloud hyperscaler, providing enterprises with cutting-edge infrastructure that keeps their data secure, their operations compliant, and their environmental impact minimal. As regulatory, environmental, and technological forces converge, the company aims to become the default platform for enterprises seeking high-performance, compliant, and sustainable AI compute in Europe.”
With this new investment, the company will expand its infrastructure, team, and service offerings with the deployment of NVIDIA’s latest-generation systems. DataCrunch has also submitted a proposal to the European Commission for the development of an AI gigafactory dedicated to AI model training and inference in Latvia.
In collaboration with the Republic of Latvia and international investor partners, DataCrunch has submitted a formal expression of interest to the European Commission for the development of an AI gigafactory dedicated to AI model training and inference, deploying around 100,000 AI accelerators.
The facility will focus on ensuring secure, regulation-compliant AI compute capacity within EU jurisdiction, providing critical compute resources for startups, SMEs, and research institutions across the EU, and reportedly utilising up to 100% renewable energy. It will be situated in Latvia, with a project structure that accommodates the inclusion of additional sites across the EU, as well as welcoming additional parties to join the initiative.
“DataCrunch exemplifies the type of European tech leadership that’s essential today,” said Henrik Rosendahl, board member in DataCrunch. “As AI becomes increasingly integral to business operations, European enterprises are seeking infrastructure partners who not only grasp complex regulatory landscapes but also share their sustainability values. DataCrunch delivers on both, while providing access to the most advanced AI compute on the market.”
Europe’s growing AI market – projected to reach $1.4 trillion by 2033 – faces a critical infrastructure challenge. US tech giants Amazon, Google and Microsoft dominate nearly 70% of the European cloud market, whilst the continent’s largest provider accounts for just 2%, creating a growing dependency on non-European hyperscalers and raising concerns around data security, rising costs, and regulatory compliance.
DataCrunch looks to address this imbalance by building a secure, sovereign, and scalable AI cloud rooted entirely in Europe.
DataCrunch presents itself as one of the few viable competitors to US-based hyperscalers, addressing growing demand from European enterprises for access to advanced hardware without compromising on data locality or sustainability.
Recent releases include:
- Instant Clusters for distributed AI workloads, which were awarded bronze in The GPU Cloud ClusterMAX Rating System by the industry analysts SemiAnalysis – matching Google Cloud Platform and surpassing 100+ other providers included in the evaluation.
- Serverless Containers enabling high-scale, low-latency AI inference at scale – handling hundreds of millions of requests per month.
Looking ahead, DataCrunch is extending its commitment to equipping the leading AI innovators with a one-stop AI development platform.
- The forthcoming releases include: Managed Kubernetes for cluster management, geographically-distributed Object Storage, and Managed Inference Endpoints with custom acceleration for the state-of-the-art generative models.
- A wide range of improvements will enhance already existing capabilities, such as Identity Access Management (IAM) control, API, cold starts of Serverless Containers, wider support of OS images, and Network Automation.
Read the orginal article: https://www.eu-startups.com/2025/09/finlands-datacrunch-raises-e55-million-to-challenge-us-hyperscalers-and-accelerate-europes-ai-sovereignty/
