Search
IDCNOVA

IBM and Together AI Sign $240M Nvidia HGX B300 Cloud Deployment Deal

By: IDCNOVARegion: North America
IBM has entered into a $240 million multi-year agreement with Together AI to deploy a large cluster of Nvidia HGX B300 systems on IBM Cloud, marking one of the most significant cloud-based AI infrastructure deals of the year. The cluster, interconnected by Nvidia Spectrum-X Ethernet technology, is expected to become available from Q1 2027 and represents the first large-scale inference-focused deployment of HGX B300 systems on IBM Cloud.

The agreement underscores the intensifying race among cloud providers to secure high-performance GPU capacity as enterprises scale their AI workloads. Together AI, a self-described AI acceleration cloud company, said it selected IBM and Nvidia for their ability to deliver GPU capacity at the pace required for rapid AI scaling and lowest token cost. The company, founded in 2022, both leases chips from other cloud providers and re-leases them to developers, while also purchasing its own servers to rent out from its data centers, which are located in North America.

Vipul Ved Prakash, CEO of Together AI, emphasized the strategic importance of the deployment. "Enterprises want the performance of the best frontier models without the closed-model price tag, and that only works if the infrastructure underneath is fast and reliable at scale," he said. "Working alongside IBM with Nvidia gives us that foundation. This cluster lets us bring production-grade inference to more companies, faster, and it's a big step in our push to make open-source AI the obvious choice for enterprises."

Alan Peacock, general manager of IBM Cloud, framed the deal as part of a broader enterprise shift toward agentic AI. "Enterprises are in a race to adopt agentic AI at scale to drive real business outcomes," he said. "IBM and Nvidia are delivering scalable, economical, enterprise-grade AI infrastructure that can help Together AI accelerate innovation for the next generation of AI infrastructure."

The agreement follows a period of rapid expansion for Together AI. In July 2026, the company raised $800 million in a Series C funding round, achieving an $8.3 billion post-money valuation. A month earlier, it signed a multi-year cloud capacity agreement with AI compute provider Rumble Inc., which will also supply Together AI with Nvidia HGX B300 systems as dedicated cloud capacity. These moves highlight Together AI's aggressive push to build out its own infrastructure footprint while maintaining flexibility through partnerships.

Industry analysts see the IBM-Together AI deal as a signal that inference workloads are becoming a primary driver of data center demand, distinct from the training-focused builds that dominated earlier AI infrastructure cycles. The use of Nvidia Spectrum-X Ethernet, rather than InfiniBand, also points to a growing preference for Ethernet-based networking in AI clusters, potentially reshaping how large-scale AI systems are architected in the coming years. For IBM, the agreement strengthens its position as a serious contender in the AI cloud market, competing with hyperscalers and specialized GPU cloud providers alike.