Axelera AI, based in Eindhoven, has secured up to €61.6 million in funding from the EuroHPC Joint Undertaking (JU) and member states as part of the DARE Project. The company aims to contribute to the development of a European supercomputing ecosystem with its AI hardware acceleration technology. This funding will be used to advance Titania, an AI inference chiplet designed for high-performance computing. The growing demand for AI-driven solutions has pushed companies to focus on scalable and energy-efficient computing architectures. Axelera AI, with its existing experience in AI hardware, seeks to address these challenges.
Axelera AI has previously received significant investments, including a $68 million Series B financing round, bringing its total funding to over $200 million in three years. The company has been working on AI inference solutions, including the Metis AI platform, which focuses on edge computing. While its earlier developments centered on computer vision applications, Titania expands its reach into broader AI workloads and high-performance computing environments. The company continues to scale its technology, aligning with Europe’s ambition to strengthen its digital sovereignty through domestic AI and supercomputing innovations.
What will the funding support?
The investment will be allocated to the development of Titania, an AI inference chiplet utilizing Axelera AI’s Digital In-Memory Computing (D-IMC) architecture. This technology allows for high scalability across computing environments, from edge devices to large-scale data centers. The company aims to improve computational efficiency while addressing concerns regarding power consumption and cooling requirements.
“Our D-IMC technology leverages a future-proof, scalable multi-AI-core architecture, ensuring unparalleled adaptability and efficiency,” said Evangelos Eleftheriou, CTO and co-founder of Axelera AI.
This approach integrates RISC-V vector extensions, enabling dynamic performance optimization across industries that require high-performance AI processing.
How will Titania impact AI infrastructure?
Titania is planned for deployment in 2028 and is expected to enhance AI throughput and efficiency in high-performance computing centers. The chiplet-based architecture, combined with RISC-V capabilities, is designed to meet evolving AI computational needs across various sectors, including robotics, data centers, and automotive industries. The design allows multiple chiplets to be integrated into a System-in-Package (SiP), optimizing performance while maintaining lower energy consumption.
“This is an important milestone and validation of our technology,” said Fabrizio Del Maffeo, co-founder and CEO of Axelera AI. “We have continuously delivered technologies to help customers tackle the AI industry’s biggest challenges and efficiently implement AI capabilities into their products.”
This step aligns with the broader European strategy of creating homegrown AI solutions that reduce reliance on non-European hardware manufacturers.
Axelera AI has its headquarters at the AI Innovation Center in Eindhoven, with additional research and development sites in Belgium, Italy, and the UK. The company employs over 200 professionals specializing in AI hardware and software, many of whom have experience at leading AI firms and multinational technology companies. By expanding its research teams in Europe, Axelera AI aims to reinforce its position in the competitive AI market.
Funding initiatives such as the DARE Project reflect Europe’s strategic push to enhance its AI and supercomputing capabilities. The EuroHPC JU’s objective of developing a unified European supercomputing infrastructure aligns with Axelera AI’s efforts to advance AI inference technologies. With increasing concerns about AI model scalability, power efficiency, and computational costs, companies working on innovative architectures like D-IMC and RISC-V are expected to play a critical role in future AI deployment strategies.
