IBM, Together AI team to offer open-source AI inference in the cloud
IBM has inked a $240 million deal with neocloud provider Together AI to offer a cloud service that will run large capacity, open-source AI inference workloads. The service will be based on a large cluster of Nvidia HGX B300 systems running Spectrum-X Ethernet networking and…
The service will be based on a large cluster of Nvidia HGX B300 systems running Spectrum-X Ethernet networking and residing in IBM Cloud. Together AI will then use this cluster to provide open-source model inference services for enterprise customers.
The Together AI platform supports open models such as DeepSeek, Nemotron, MiniMax, Kimi, and GLM that the company says gives developers and customers the freedom to customize and fine-tune them at a lower-cost than a single proprietary model for their own applications. Together AI.
What Happened
A hybrid environment on IBM Cloud powered by Nvidia GPUs connected with Nvidia Spectrum-X Ethernet networking and Together AI’s inference platform gives organizations a reliable foundation to build, deploy and scale AI systems, IBM stated. In addition, the technology behind the agreement.
IBM has inked a $240 million deal with neocloud provider Together AI to offer a cloud service that will run large capacity, open-source AI inference workloads.
Gartner just this week forecast that worldwide spending on AI-optimized IaaS is projected to grow 96% through 2026, reaching $42 billion.
The market will sustain high growth and reach $66 billion in 2027, according to the firm.
Key Details
The two vendors recently expanded their collaboration to grow the use of GPU-native data analytics, intelligent document processing, on-premises and regulated infrastructure deployments to help enterprise customers move AI from pilot to production, according to IBM. Vipul Ved Prakash, CEO at Together.
In 2026, global spending on inference ($23.3 billion) will surpass that of training ($19 billion), Gartner stated.
Fifty-five percent of AI-optimized IaaS spending is forecast to support inference in 2026 and is set to reach 59% in 2027.
He has a BA in Journalism and Technical Writing Certificate from The Pennsylvania State University.
Why It Matters
The deal is the latest to contribute to a swell of AI-optimized infrastructure as a service (IaaS) activity. IBM Cloud competes with AWS, Google Cloud, Microsoft Azure, Coreweave, Lambda and many others in a rapidly expanding AI IaaS market.
Michael has been writing about the industry for more than 33 years and has won coverage awards from ASBPE and FOLIO.
Michael is a senior editor with Network World focused on deciphering the strategies of many core high-tech vendors such as Cisco, Arista, Juniper, HPE and IBM.
What Reports Say
Coverage of the story so far points to:
Continued reporting by Network World as more details emerge
