IBM is planning to make AMD Instinct MI300X GPUs available as a service via IBM Cloud.
The GPUs will be rolled out on the platform in the first half of 2025, with the aim of improving the performance and power efficiency of generative AI models and high-performance computing applications.
“As enterprises continue adopting larger AI models and datasets, it is critical that the accelerators within the system can process compute-intensive workloads with high performance and flexibility to scale,” said Philip Guido, executive vice president and chief commercial officer, AMD.
“AMD Instinct accelerators combined with AMD ROCm software offer wide support including IBM watsonx AI, Red Hat Enterprise Linux AI, and Red Hat OpenShift AI platforms to build leading frameworks using these powerful open ecosystem tools. Our collaboration with IBM Cloud will aim to allow customers to execute and scale Gen AI inferencing without hindering cost, performance, or efficiency.”
To be available via IBM Cloud, the MI300X GPUs have 192GB of high-bandwidth memory meaning they can run larger models with few GPUs.
The MI300X was launched in December 2023. IBM is the latest in a series of companies to make the hardware available to customers.
September saw both Oracle Cloud Infrastructure and Vultr adding the GPUs to their offerings. Microsoft is also adding the GPU to its platform. GPU cloud provider Nscale is specifically offering AMD hardware, including the MI300Xs.
IBM Cloud added Nvidia H100 GPUs to its cloud platform in October 2024, a year after deploying Nvidia A100s. It is also adding Intel's Gaudi 3 to its offerings.
IBM is reportedly in discussions with AWS for a $475m deal in which IBM will use AWS for AI cloud services.
