IBM has announced the upcoming general availability of the IBM Spyre Accelerator, a cutting-edge AI accelerator designed to deliver low-latency inferencing for generative and agentic AI applications. This innovation aims to empower enterprises to scale AI workloads while ensuring the security and resilience of critical systems.
Designed for Modern AI Demands
As enterprises increasingly adopt AI technologies, the need for infrastructure that can handle the demands of generative and agentic AI becomes paramount. IBM’s Spyre Accelerator addresses this need by providing a solution that supports AI inferencing without compromising the integrity of core data and applications. This enables organizations to keep mission-critical data on-premises, thereby mitigating risks and enhancing operational efficiency.
A Product of Extensive Research and Collaboration
The Spyre Accelerator is the culmination of years of research and development at IBM’s AI Hardware Center. Initially introduced as a prototype, the accelerator underwent rigorous testing and refinement, including cluster deployments at IBM’s Yorktown Heights campus and collaborations with institutions like the University at Albany’s Center for Emerging Artificial Intelligence Systems. These efforts have resulted in a commercial-grade system-on-a-chip featuring:
-
32 individual accelerator cores
-
25.6 billion transistors
-
5nm node technology
-
Mounted on a 75-watt PCIe card
This design allows for clustering up to 48 cards in IBM z17 or LinuxONE systems and up to 16 cards in IBM Power systems, enabling scalable AI capabilities.
Also Read: AMD & OpenAI Partner to Deploy 6 Gigawatts of AMD GPUs
Strategic Insights from IBM Leadership
Barry Baker, Chief Operating Officer of IBM Infrastructure and General Manager of IBM Systems, emphasized the strategic importance of the Spyre Accelerator:
“One of our key priorities has been advancing infrastructure to meet the demands of new and emerging AI workloads. With the Spyre Accelerator, we’re extending the capabilities of our systems to support multi-model AI – including generative and agentic AI. This innovation positions clients to scale their AI-enabled mission-critical workloads with uncompromising security, resilience, and efficiency, while unlocking the value of their enterprise data.”
Mukesh Khare, General Manager of IBM Semiconductors and Vice President of Hybrid Cloud, highlighted the foundational work behind the accelerator: “We launched the IBM Research AI Hardware Center in 2019 with a mission to meet the rising computational demands of AI, even before the surge in LLMs and AI models we’ve recently seen. Now, amid increasing demand for advanced AI capabilities, we’re proud to see the first chip from the Center enter commercialization, designed to deliver improved performance and productivity to IBM’s mainframe and server clients.”
Enhanced AI Capabilities with Telum II Processor
In IBM mainframe systems, the Spyre Accelerator is paired with the Telum II processor, offering enhanced security, low latency, and high transaction rate processing power. This combination allows businesses to scale multiple AI models simultaneously, including large language models (LLMs), within a trusted environment.