Microsoft Azure launches world’s first Nvidia GB300 cluster for OpenAI

Microsoft Azure has unveiled the large-scale NDv6 GB300 VM series, considering the world’s first supercomputing cluster of Nvidia GB300 NVL72 system for OpenAI’s most challenging AI interface.

The supercomputer cluster is powered by over 4600 Nvidia Blackwell Ultra GPUs which are connected through Quantum-X800 InfiniBand networking.

NDv6 GB300 VM is the liquid-cool and rack-scale system. Being a powerhouse, each rack integrates 72 Nvidia Blackwell Ultra GPUs and 36 Nvidia Grace CPUs into a single cohesive unit to speed up interface and training for models.

The system also features 37 terabytes of fast memory and 1.44 exaflops of FP4 Tensor Core performance per VM, offering a staggering memory space f0r agentic AI systems.

This cutting-edge system consists of radical engineering to networking and memory, providing a massive scale of computing to achieve high interface functionality.

The deployment of the supercomputing cluster uplifts OpenAI’s artificial intelligence models by reducing training duration from weeks to days.

The launch also highlights the partnership between Microsoft and Nvidia for expanding next-generation AI infrastructure.

According to Nidhi Chappell, corporate vice president of Microsoft Azure AI Infrastructure, “Delivering the industry’s first at-scale NVIDIA GB300 NVL72 production cluster for frontier AI is an achievement that goes beyond powerful silicon — it reflects Microsoft Azure and NVIDIA’s shared commitment to optimize all parts of the modern AI data center.”

“Our collaboration helps ensure customers like OpenAI can deploy next-generation infrastructure at unprecedented scale and speed,” she added.