Kalray Announces the Availability of Its TURBOCARD4 Card: a Paradigm Shift in Compute Acceleration for Smart Vision and Gen-AI Data-indexing Applications

Kalray (Euronext Growth Paris: ALKAL), a leading provider of hardware and software solutions to accelerate data intensive workflows from Cloud to Edge announces the commercial availability of its latest product, the TURBOCARD4 (TC4) acceleration card. Powered by four of Kalray’s latest generation of DPU processors, the TC4 targets the fast growing markets of AI-powered smart vision and data-indexing applications. The TC4 is manufactured in France, in the French plant of Asteelflash, one of the leading global manufacturers worldwide. Kalray already received an initial order of more than $1 million (hundreds of cards)¹ to be recognized this year.

TC4 TARGETS THE FAST GROWING COMPUTE ACCELERATION CARD MARKET

Whereas enterprises and data centers have to process an ever-growing amount of data, compute acceleration cards are getting more and more traction to bring the best performance and best power efficiency in data processing.

“The availability of our new TC4 acceleration card is a major step for Kalray to set the stage for the widespread adoption of our latest acceleration technology. The TC4 has already been selected by one of our major customers and we envision a fast commercial take-off due to the ever-increasing demand in data processing acceleration solutions”, shares Eric Baissus, Kalray CEO.

“In addition, we are very proud to announce that the TC4 is our first “made in France” product. The production is ramping up rapidly in the plant of the leading global manufacturer AsteelFlash in Langon”.

BUILT TO COMBINE AI AND CLASSICAL COMPUTING

TC4 embodies Kalray's vision in addressing the intricate demands of modern computing workloads. Housing four of the latest generation of Kalray’s DPUs, Coolidge2^TM, in a single PCIe card, Kalray’s TC4 is designed to ensure customers can merge classical and AI-based processing technologies and create superior, efficiency-driven systems for the most processing-intensive AI applications. DPUs offer a very complementary architecture to GPUs, allowing for the processing of a large number of different operations in parallel in an asynchronous way. DPUs are well-suited for pre-processing data that is later used by GPUs or in the context of complex intelligent systems running many different algorithms in parallel.

TWO FAST GROWING MARKETS TARGETTED: THE SMART VISION MARKET…

TC4 is strategically targeted at two rapidly growing markets. The first one is the Smart Vision market. Smart vision applications play a pivotal role in transformative sectors such as Industrial, Health Science, or Smart City operations due to their ability to analyze visual data in real-time, make intelligent decisions, and automate complex processes. The smart vision market size was valued at USD 20.31 billion in 2023 which is forecasted to reach USD 175.72 billion by 2032, exhibiting a CAGR of 27.3% ​ (Fortune Business Insights)​. These projections highlight the increasing demand of smart vision technologies across various industries, driven by advancements in AI and machine learning, and their application in automation, quality control, and enhanced operational efficiencies.

…AND THE DATA-INDEXING FOR GEN-AI MARKET

In the context of Generative AI, data indexing is crucial for several reasons. Generative AI models, such as those used in natural language processing or image generation, often need to access enormous datasets. Efficient indexing minimizes the time required to fetch the relevant data, making the training and inference processes faster.

In addition, as Generative AI models become more complex, the datasets they train on grow exponentially. Data indexing allows these models to scale more effectively by ensuring that the increase in data volume does not linearly increase access times.

Data-Indexing also improves models’ outputs accuracy as indexing helps quickly identify the most relevant data pieces to insert in model context and prompt. This ensures that the model's outputs are as accurate and contextually appropriate as possible.

TC4 PRODUCTION HAS BEEN RE-LOCATED TO ASTEELFLASH FRENCH PLANT

Kalray's commitment to quality and desire to re-locate its production to Europe led to the alliance with Asteelflash for TC4 production. Asteelflash is one of the leading global electronic manufacturing services (EMS) companies, specializing in complex and high-tech manufacturing solutions across various industries. Asteelflash has several plants in Europe, including Germany and France. The TC4 production is performed in the Langon, France plant. This partnership has been supported by the French government as part of the CARAIBE funded project. The production volume is expected to grow from a few hundred per month today to several thousands of units per month, starting in 2025.

ABOUT KALRAY

Kalray is a leading provider of hardware and software technologies and solutions for high-performance, data-centric computing markets, from cloud to edge.

Kalray provides a full range of products to enable smarter, more efficient, and energy-wise data-intensive applications and infrastructures. Its offers include its unique patented DPU (Data Processing Unit) processors and acceleration cards as well as its leading-edge software-defined storage and data management offers. Separated or in combination, Kalray’s high-performance solutions allow its customers to improve the efficiency of data centers or design the best solutions in fast-growing sectors such as AI, Media & Entertainment, Life Sciences, Scientific Research, Edge Computing, Automotive and others.

Founded in 2008 as a spin-off of the well-known French CEA research lab, with corporate and financial investors such as Alliance Venture (Renault-Nissan-Mitsubishi), NXP Semiconductors or Bpifrance, Kalray is dedicated through technology, expertise, and passion to offer more: more for a smart world, more for the planet, more for customers and developers. www.kalrayinc.com

¹ See press release as of 11th January, 2024

View source version on businesswire.com: https://www.businesswire.com/news/home/20240402257071/en/