Axelera AI Secures $68M to Challenge NVIDIA in Edge AI Chip Market

Axelera AI raises $68M in Europe’s largest Series B round in the fabless semiconductor category to rival NVIDIA with its edge AI chips.

Key details of Axelera’s funding round: The Netherlands-based startup has secured $68 million in Series B funding, bringing its total capital raised to $120 million, to expand its AI processing unit (AIPU) solutions:

• The funding was led by major institutional investors including Invest-NL Deep Tech Fund, European Innovation Council Fund, Innovation Industries Strategic Partners Fund and Samsung Catalyst Fund.
• Axelera plans to use the capital to bring its existing AIPU solutions to new geographies and markets, while also developing new products to address the computing needs of next-generation AI workloads like multimodal large language models (LLMs).

Axelera’s edge AI acceleration platform: The company’s Metis platform takes a hardware and software approach to enable efficient computer vision inference at the edge:

• At the core is a 12nm CMOS AIPU chip with four “self-sufficient” AI cores that can collaborate on workloads or process different neural networks concurrently, delivering up to 214 TOPS of AI processing power.
• Axelera’s proprietary digital in-memory computing technology accelerates matrix operations, offering high energy efficiency at 15 TOPS/W by using memory devices to store matrices and perform in-place computations without data movement.
• A software development kit allows enterprises to easily build computer vision applications that run on devices using the Metis AIPU.

Growth plans and market expansion: With strong demand and a $100M+ business pipeline, Axelera is focused on scaling up production and expanding into new verticals:

• The company is currently shipping Metis evaluation kits to early customers and plans to ramp up to full production for widespread deliveries.
• Key target markets include automotive, digital healthcare, Industry 4.0, retail, robotics, and surveillance across North America, Europe and the Middle East.
• Beyond edge AI, Axelera also aims to launch cost-effective data center accelerators to address the growing compute requirements of generative AI models.

Broader context in the AI hardware landscape: As NVIDIA continues to dominate the AI chip market, Axelera is among a group of startups carving out niches in specific domains like edge computing:

• Axelera CEO Fabrizio Del Maffeo emphasized the need for high-performance, efficient and cost-effective solutions to “democratize access to artificial intelligence, from the edge to the cloud.”
• Other notable players in this space include Hailo, which recently unveiled its Hailo-10 processor designed for deploying generative AI applications on edge devices like cars and robots.
• With major funding rounds and new products, these startups are aiming to challenge NVIDIA’s dominance and address the evolving landscape of AI workloads across different computing environments.