Both companies said the verification is designed to advance the development of functionality to “partition and properly allocate GPU resources according to requirements” – which includes AI application model size and the number of concurrent executions.
This will take place through SoftBank enhancing its mechanism for managing computing resources as well as optimally allocating AI applications, known as Orchestrator.
“Through joint validation with AMD, we implemented orchestration logic to control AMD Instinct GPUs,” said Ryuji Wakikawa, vice president and head of the Research Institute of Advanced Technology at SoftBank Corp. “This enables more efficient operation of multiple AI applications on a single GPU. SoftBank will continue to improve the efficiency of computing resource utilisation.”
The news comes as interest in generative AI (Gen AI) continues to surge, prompting rising demand for applications powered by large language models (LLMs). LLMs require computing resources that significantly vary, which means allocating resources uniformly can lead to a shortage or surplus of a GPU – potentially resulting in operation being inefficient.
SoftBank has been developing Orchestrator amid this backdrop. The tool manages computing resources and optimally allocates AI applications with the goal of achieving an AI infrastructure that can flexibly control computing resources according to specific requirements.
With AMD, the company has developed an enhanced Orchestrator feature that leverages the GPU partitioning capabilities of AMD Instinct GPUs. This enables a single GPU to be used as multiple logical devices, therefore enabling flexible allocation of GPU resources.
The companies said this makes it possible to run multiple AI applications on a single GPU. It has also been confirmed that this enhanced Orchestrator is able to operate with minimal strain on GPU resources and occurrence of surplus.
“AMD and SoftBank share a focus on delivering high-performance, efficient infrastructure that helps bring AI inference into real-world deployments. Through our collaboration, SoftBank is right-sizing GPU resource allocation to match model requirements and help enable flexible inference platforms that support a wide range of real-world AI services,” said Kumaran Siva, corporate vice president, strategic market development at AMD.
“We are excited to continue working with SoftBank to advance enterprise technologies AI inference for next generation AI infrastructure.”
Related stories
25 years of Metro Connect USA: Your complete event guide
AI chip crunch continues, as Intel and AMD confront CPU delays
Metro Connect USA 2026
Metro Connect USA is the largest executive-level digital infrastructure event in the U.S. The only one of its kind, this 25-year-strong gathering is where decision makers come together to make deals happen.
