IBM and Together AI Sign Multi-Year Agreement to Scale Open-Source AI Inference with NVIDIA AI Infrastructure on IBM Cloud - PR Newswire

IBM and Together AI have entered a multi-year agreement to scale open-source AI inference using NVIDIA AI infrastructure on IBM Cloud. Under the $240 million deal, IBM will deploy a large-scale infere... Read original →

This collaboration signals a major leap in enterprise AI inference capabilities, with NVIDIA's HGX B300 and IBM Cloud forming a powerful synergy that dramatically improves large-scale model throughput, especially in token economics and compute cost efficiency. The move directly stimulates upstream GPU demand, accelerating NVIDIA’s market penetration in enterprise AI infrastructure. Downstream, it accelerates the commercialization of open-source AI models, reinforcing Together AI’s platform leadership in inference, training, and agentic workflows. From a compliance standpoint, this partnership intensifies U.S.-China tensions in AI compute, likely triggering stricter export controls. Competitors like Google, AWS, and Microsoft will respond with accelerated self-developed chip and cloud integrations. Over the next 12–24 months, the AI factory model will dominate enterprise adoption, driving a surge in demand for high-throughput, low-latency inference clusters—marking the onset of a new wave of AI infrastructure investment.