IBM Corp. and Together AI have entered a multi-year, $240 million agreement to deploy a large-scale cluster of NVIDIA Corp. HGX B300 systems on IBM Cloud.
The strategic collaboration announced Tuesday is designed to enhance open-source artificial intelligence (AI) model inference and provide cost-effective scaling for enterprise AI operations.
Under terms of the deal, IBM will build its first dedicated, large-scale inference cluster using the NVIDIA B300 hardware integrated with NVIDIA Spectrum-X Ethernet networking. The combined architecture is expected to increase AI factory output by up to 30 times compared to previous generations, delivering optimized token economics and performance for developers and corporate clients.
Together AI will utilize the IBM Cloud infrastructure to support its rapidly expanding inference platform, which currently processes 400 trillion tokens per month. The firm, which recently closed an $800 million Series C funding round at an $8.3 billion valuation, focuses on open-source AI models across inference, training, fine-tuning, and agentic workflows.
Leadership from all three organizations framed the partnership as a critical step toward establishing scalable, enterprise-grade AI infrastructure.
“Together AI is proud to lead the way in bringing production inference to market with NVIDIA’s latest AI infrastructure on IBM Cloud,” Together AI CEO Vipul Ved Prakash said in a statement. Prakash noted that working with IBM and NVIDIA aligns with the company’s goal of making advanced open-source AI broadly accessible through improved infrastructure and inference economics.
The agreement comes as enterprise demand accelerates for agentic AI applications capable of handling complex, real-time tasks.
“Enterprises are in a race to adopt agentic AI at scale to drive real business outcomes,” said Alan Peacock, general manager of IBM Cloud, emphasizing that the joint solution provides the necessary economic efficiency and scale.
Dion Harris, senior director of HPC and AI Infrastructure Solutions at NVIDIA, highlighted the shift toward continuous compute environments.
“AI factories are becoming essential enterprise infrastructure—like electricity and telecommunications—turning compute and data into intelligence,” Harris said, adding that the platform gives enterprises the efficiency required for real-time services.
The deployment represents an expansion of IBM and NVIDIA’s broader joint initiatives, which cover hybrid cloud computing, GPU-native data analytics, unstructured data processing, and consulting services designed to help organizations operationalize AI.