TLDR: IBM and Intel have partnered to bring Intel Gaudi 3 AI accelerators to IBM Cloud, aiming to provide a cost-effective and high-performance solution for enterprise AI workloads. This collaboration, announced at Intel Vision 2025, makes IBM Cloud the first global cloud service provider to offer Gaudi 3, with initial availability in Frankfurt and Washington, D.C., and further expansion planned. The integration is designed to support generative AI, model fine-tuning, and inferencing, offering flexible deployment options and aiming to address the rising costs associated with AI infrastructure.
IBM and Intel have announced a significant collaboration to deploy Intel Gaudi 3 AI accelerators as a service on IBM Cloud, marking a pivotal step in making advanced AI capabilities more accessible and cost-effective for enterprises. This strategic partnership, unveiled at Intel Vision 2025, positions IBM Cloud as the inaugural global cloud service provider to offer Intel Gaudi 3, with initial availability commencing in March 2025 in the Frankfurt (eu-de) and Washington, D.C. (us-east) IBM Cloud regions. Further expansion to the Dallas (us-south) IBM Cloud region is anticipated in Q2 2025.
The integration of Intel Gaudi 3 AI accelerators is specifically designed to cater to high-performance AI workloads, including generative AI inferencing, model fine-tuning, and the processing of large language models (LLMs) and multi-modal models (MMMs). The Gaudi 3 accelerators boast impressive specifications, featuring 64 Tensor Processor Cores (TPCs) and eight Matrix Multiplication Engines (MMEs) to accelerate deep neural network computations. They are equipped with 128 GB of HBM2E memory, offering up to 3.7 TB/s of memory bandwidth, and support industry-standard Ethernet networking with 24×200 GbE ports, providing 9.6 Tbps of bi-directional bandwidth for scalable system interconnectivity. These accelerators are paired with 5th Gen Intel Xeon processors on IBM Cloud Virtual Servers for VPC.
According to IBM’s “AI in Action 2024” report, 67% of surveyed business leaders reported revenue increases of 25% or more due to the inclusion of AI in their business operations. However, the report also highlighted the challenge of balancing these gains with the infrastructure costs required to drive AI performance. This collaboration directly addresses this concern by aiming to provide better cost performance for AI innovation.
Saurabh Kulkarni, Vice President, Datacenter AI Strategy and Product Management at Intel, emphasized the significance of this partnership, stating, “By bringing Intel Gaudi 3 AI accelerators to IBM Cloud, we’re enabling businesses to help scale generative AI workloads with optimized performance for inferencing and fine-tuning. This collaboration underscores our shared commitment to making AI more accessible and cost-effective for enterprises worldwide.” Justin Hotard, Intel Executive Vice President and General Manager of the Data Center and AI Group, added, “By integrating Gaudi 3 AI accelerators and Xeon CPUs with IBM Cloud, we are creating new AI capabilities and meeting the demand for affordable, secure and innovative AI computing solutions.”
Businesses will have flexible deployment options for integrating Gaudi 3 into their AI infrastructure. These include standalone servers via IBM Cloud Virtual Servers for VPC, which are ideal for users requiring full control over their software stacks and support Red Hat Enterprise Linux AI images. IBM also plans to integrate Gaudi 3 into Red Hat OpenShift AI clusters in Q2 2025 for containerized environments. Furthermore, enterprises will be able to run IBM watsonx.ai, IBM’s AI development studio, directly on Gaudi 3-powered virtual servers, offering comprehensive control over the entire AI stack. Support for IBM watsonx, Red Hat OpenShift AI clusters, IBM Cloud Kubernetes Service, deployable architectures on IBM Cloud, and an automated Terraform-based deployment are planned for the second half of 2025.
Also Read:
- IBM Unveils Advanced AI Integration Services to Empower Enterprises in Building and Scaling AI Agents
- Lenovo Leverages AI to Streamline Sales and Drive Business Growth
This collaboration is expected to provide a scalable cloud alternative for enterprises, promising enhanced control over AI workloads, improved security, and a compelling balance of low costs and high performance.
