Euclyd raises €200+ million / $231 million Series A co-led by Samsung and Somerset Capital for AI inference chip development

Euclyd, a Dutch semiconductor startup founded in 2024, has raised $231 million (200 million euros) in a Series A round co-led by Samsung alongside Somerset Capital Partners, the Scaleup Europe Fund managed by EQT, and Innovation Industries. The company, based at High Tech Campus Eindhoven in the Netherlands and founded by Bernardo Kastrup and Atul Sinha, is building chip systems designed to reduce the energy consumption and cost of running AI models.

Euclyd's focus is on AI inference—the stage at which a trained model responds to queries—rather than the computationally heavier model training workload. The company's hardware combines programmable compute, processor-memory co-design, and system-level optimization. On the commercial side, Euclyd pursues two distinct business lines: direct sales of rack-based hardware to enterprises running AI inference on their own infrastructure, and licensing arrangements with chip developers who want to build on Euclyd's underlying technology.

Chief executive Bernardo Kastrup emphasized Samsung's strategic value beyond capital. "They are one of the biggest memory manufacturers in the world. They do a lot of engineering, they know a lot about systems, they know the supply chain, they have a huge network," he said.

Dede Goldschmidt, senior vice president of Samsung Electronics and head of the Samsung Semiconductor Innovation Center, framed the investment within the broader trajectory of AI infrastructure. "The next phase of AI will be defined not only by model innovation, but also by the efficiency and scalability of its infrastructure," Goldschmidt said.

Euclyd plans to use the proceeds to expand its engineering team, advance silicon and systems development, and build ecosystem partnerships. The company expects to launch its physical chip systems in 2028 and serve thousands of enterprise customers by 2030, though it has not yet demonstrated its systems in large-scale commercial environments.

The funding reflects growing investment in alternatives to Nvidia's dominance. Nvidia's share of the inference chip segment is estimated between 60 and 75 percent in 2025, but faces mounting pressure from custom silicon developed by hyperscalers and startups. Inference—where deployed models handle live queries—is where competition is most active, as challengers have found Nvidia's market position there comparatively weaker than in training workloads.

Peter Wennink, chair of Euclyd's board and former president and chief executive of ASML, voiced confidence in Europe's competitive positioning. "EUCLYD can transform those strengths into a globally competitive AI infrastructure platform," he said.