IBM, Together AI Partner on $240M Open-Source AI Inference Deal
IBM and Together AI have partnered to provide cost-effective open-source AI inference for enterprise customers. The two companies announced a multi-year commercial agreement worth $240 million in August 2026, which will power Together AI's infrastructure on IBM Cloud.
The collaboration aims to increase efficiency and token economics for enterprises by providing high-performance infrastructure optimized for scaling open-source AI model inference. To achieve this, IBM will deploy an NVIDIA HGX B300 cluster on IBM Cloud, utilizing the NVIDIA Spectrum-X Ethernet interconnected system for high-performance communication between nodes.
The large-scale computer cluster is scheduled to be generally available in Q1 2027 and will be delivered through IBM Cloud. According to Together AI CEO Vipul Ved Prakash, the goal is to provide enterprise users with performance comparable to closed frontier models without the high costs of proprietary APIs, lowering the cost per token for production workloads.