Eindhoven-based Axelera AI has announced the close of its 27-million-dollar series-A investment round. The funding will support the launch and mass production of the European startup’s first-generation artificial-intelligence acceleration platform, powered by its in-memory and dataflow technologies. The money is also earmarked for further team expansion. The company will shortly announce its first product, set to launch to select customers and partners in early 2023, and begin raising for its series B later this year.
The series-A round was led by Innovation Industries and joined by Imec.xpand and the Federal Holding and Investment Company of Belgium (SFPIM). In addition, the Netherlands Enterprise Agency (RVO) awarded Axelera AI a 6.7-million-dollar Innovation Credit, a special loan reserved for highly innovative and impactful projects, commissioned by the Dutch Ministry of Economic Affairs and Climate Policy. Innovation Industries and Imec.xpand previously participated in the company’s $12M seed round.
Jonathan Ballon, former vice president and general manager of Intel’s Edge AI and IoT business, has come onboard as Axelera’s chairman. He joins the non-executive board, alongside Cyril Vančura, partner at Imec.xpand, Sander Verbrugge, partner at Innovation Industries, and Valery Vavilov, founder and CVO at Bitfury Group. Marc Ceurremans from the SFPIM will serve as board observer.

Incubated by Bitfury in 2019, Axelera got its start operating as Bitfury AI, before spinning out as an independent entity in 2021. Leveraging expertise from IBM, Imec, Intel and Qualcomm, it wants to lower the AI threshold for European SMEs by supplying a cost-effective range of plug-and-play products based on its own silicon. Its game-changing hardware and software solution will concentrate the AI computational power of an entire server into a single chip at a fraction of the energy consumption and price of AI hardware today.
Since its inception and initial funding last year, Axelera has successfully taped out its first testbench chip, proving the performance and efficiency of in-memory compute for AI computations. The Thetis Core was developed entirely in-house. Significant progress has been made in the software stack as well, which will allow customers to easily run their neural networks on the company’s accelerator without any model retraining.
In just fourteen months, Axelera has hired more than eighty team members, distributed between its headquarters at the High Tech Campus in Eindhoven, its R&D offices in Leuven and Zurich, its newest offices in Milan and Bristol, and its Europe-wide remote locations. The company plans to expand to the United States and Taiwan this year.