March 06, 2024 – Eindhoven, Netherlands: AI hardware maker Axelera AI has unveiled Titania, which the company described as a high-performance, low-power and scalable AI inference chiplet. Part of the EuroHPC Joint Undertaking’s effort to develop a supercomputing ecosystem in Europe, Axelera AI said it is receiving up to €61.6 million in funding as part of EuroHPC JU’s Autonomy of RISC-V for Europe (DARE) Project.
This new funding follows the successful close of an oversubscribed $68 million Series B financing round for the company, bringing the total amount raised to more than €200 million in three years.
According to Axelera AI, the Titania chiplet will build on Axelera AI’s approach to digital in-memory computing architecture, which provides near-linear scalability from the edge to the cloud, according to the company.
As part of the DARE consortium, Axelera AI will support the EuroHPC JU and its effort to develop a supercomputing ecosystem in Europe. DARE aims to foster the design and development of European processors, accelerators, and related technologies for extreme-scale, high-performance, and emerging applications.
“Our Digital In-Memory Computing (D-IMC) technology leverages a future-proof, scalable multi-AI-core architecture, ensuring unparalleled adaptability and efficiency. Enhanced with proprietary RISC-V vector extensions, this versatile mixed-precision platform is engineered to excel across diverse AI workloads,” explained Evangelos Eleftheriou, CTO and co-founder of Axelera AI. “Uniquely, our architecture facilitates scaling from the edge to the cloud, streamlining expansion and optimizing performance in ways that traditional cloud-to-edge approaches cannot. We are setting a new standard for AI infrastructure, making true scalability a tangible reality”
Citing AI market is growth at 28 percent+ CAGR with the vast majority of that expansion driven by inference, the company said existing concerns around performance, cost, efficiency and sustainability of cloud-based solutions are intensifying due to industry advancements. Innovations such as reasoning models (i.e. OpenAI-o1 and DeepSeek R1) require significantly more inference computing than earlier transformer models. Targeting a deployment date of 2028, Axelera AI said Titania is engineered to address these challenges by delivering superior throughput and efficiency for data-intensive AI applications and future zetta-scale HPC centers at a competitive price.
To support this development effort, Axelera will grow its research and development teams in the Netherlands, Italy and Belgium.
The Titania chiplet-based architecture will leverage the D-IMC technology along with RISC-V capabilities, designed for AI demands HPC, enterprise data centers, robotics, automotive and others, while maintaining the efficiency of an edge-oriented architecture. The company said D-IMC allows for near-linear scalability without the significant power and cooling overhead typical of other solutions. Additionally, integrating RISC-V technology with vector extensions enables Axelera AI to rapidly innovate in response to evolving customer needs. Multiple Titania chiplets will be packaged in a System-in-Package (SiP).
“This is an important milestone and validation of our technology. Since Axelera AI was founded in July 2021, we have continuously delivered technologies to help customers tackle the AI industry’s biggest challenges and efficiently implement AI capabilities into their products,” said Fabrizio Del Maffeo, Co-Founder and CEO at Axelera AI. “Today, we deliver a cutting-edge hardware and software platform for accelerating computer vision on edge devices at a fraction of the cost and energy consumption of current solutions. Titania builds upon this unique product suite. We are grateful to the EuroHPC DARE Project and the countries involved for helping accelerate the development of this groundbreaking AI inference technology for HPC data centers.”
The development of Titania seamlessly aligns with Axelera AI’s mission to democratize AI and complements the company’s current product offerings. This includes the Metis AI Platform designed to simplify AI inference acceleration.
Headquartered in the AI Innovation Center of the High Tech Campus in Eindhoven, The Netherlands, Axelera AI has R&D offices in Belgium, Switzerland, Italy and the UK, with more than 200 employees across three continents.
