Euclyd Secures $230M in Samsung-Led Funding to Challenge GPU Dominance

The 20-second version

• Samsung and Somerset Capital Partners co-led the $230 million Series A round for Dutch chip designer Euclyd.
• The startup is engineering a novel architecture for AI inference to reduce operational electricity demands.
• Physical hardware delivery is projected for 2028, with a commercial goal of thousands of clients by 2030.

Why it matters

The massive capital injection into Euclyd reflects an industry-wide push to break the current GPU monopoly and establish more cost-effective, energy-efficient infrastructure for scaling AI applications.

The story

A Dutch semiconductor venture, Euclyd, has finalized a $230 million Series A financing round to accelerate the development of its proprietary hardware. The investment was co-led by Samsung alongside several European funds, including Innovation Industries and the Scaleup Europe Fund. This transaction represents a record-breaking early-stage capital raise for the Netherlands' chip sector, signaling high investor confidence in non-GPU architectures.

Established in 2024, Euclyd intends to disrupt the current market by focusing exclusively on the inference stage of artificial intelligence. While existing processors from market leaders are often repurposed gaming components, Euclyd’s silicon is being built from the ground up to minimize the financial and environmental costs of running enterprise models. The company anticipates that its specialized design will offer a competitive advantage as data centers reach their power limits.

Samsung's involvement provides more than just liquidity; the South Korean conglomerate brings manufacturing scale and memory technology integration. This strategic alliance is part of a broader trend where major tech entities are diversifying their hardware pipelines. Samsung is concurrently dedicating substantial foundry capacity to high-bandwidth memory projects for other custom accelerator initiatives, such as the Jalapeño processor and HBM4 supply chains.

The competitive landscape for AI silicon is tightening as other startups secure similar funding levels. For instance, Positron recently obtained an identical $230 million sum to advance its Atlas and Asimov chips, which also target reduced power consumption for inference tasks. These emerging players are betting that the next phase of the AI boom will prioritize sustainable operating margins over raw training power.

Euclyd faces a significant engineering runway before reaching commercial viability. The firm does not expect to ship tangible chip systems until 2028. Following that milestone, management has set a target to expand its reach to thousands of corporate users by the end of the decade. This long-term timeline highlights the capital-intensive nature of challenging established incumbents in the high-end semiconductor space.

The other side

The five-year gap before Euclyd reaches scale allows established providers to refine their own low-power offerings, potentially neutralizing the startup's architectural advantages before it ships its first unit.

What's next

Euclyd will now focus on finalizing its initial silicon specifications and leveraging Samsung’s supply chain resources to move toward its 2028 hardware launch deadline.

Sources

• CNBC Top NewsSamsung backs Nvidia AI chip rival in $230 million funding round as GPU alternatives boom
• pluang.comDutch AI chip startup Euclyd raises $230M to ri...
• app.dealroom.coSamsung backs Nvidia rival Euclyd in $230M Series A
• techcrunch.comExclusive: Positron raises $230M Series B to take on Nvidia's AI chips | TechCrunch
• aimultiple.comTop 30+ AI Chip Makers: NVIDIA & Its Competitors