07/02/2024 | Press release | Distributed by Public on 07/02/2024 00:02
Milan, 2 July 2024. DHH S.p.A. (DHH.MI) (ISIN shares IT0005203622), through its subsidiary Seeweb, announces the launch of the AI Accelerator product. This new offering is designed to enhance AI inference workloads by utilizing Tenstorrent cards, providing significant computational power and efficiency for AI applications.
About the AI Accelerator
The AI Accelerator product leverages advanced Tenstorrent technology, specifically neural processing units (NPUs), to deliver high-performance computing tailored for AI inference. This solution targets organizations requiring robust and scalable AI processing capabilities to handle complex data-driven tasks.
Key Features
Understanding Neural Processing Units (NPUs)
Neural processing units (NPUs) are specialized hardware designed to accelerate machine learning tasks, particularly in AI inference. Simplifying the difference between GPUs and NPUs, GPUs are more "general purpose" and used in various stages of AI development, while servers with NPUs are more suitable for "light machine learning" and small models, especially in inference. Additionally, GPUs incur high costs, including energy consumption, whereas NPUs offer a more economical and energy-efficient solution. The distinctive characteristics of NPUs compared to GPUs include:
How It Works
The AI Accelerator integrates seamlessly into existing systems, allowing users to enhance their AI capabilities with minimal disruption. By using state-of-the-art Tenstorrent cards, the product ensures high performance and reliability, supporting a wide range of AI applications from natural language processing to computer vision.
About Tenstorrent
Tenstorrent, headquartered in Toronto, Canada, is a VC-backed leading developer of next-generation computing hardware, specializing in designing and manufacturing neural processing units (NPUs) that accelerate AI workloads.
Market Impact
This new product is part of DHH's comprehensive efforts to develop the next-generation internet infrastructure, natively designed to support AI applications, ensuring seamless integration, scalability, and performance.
For more information about the AI Accelerator, visit Seeweb Blog.