Skip to main content
Powered by eight NVIDIA B300 Blackwell GPUs, these instances deliver outstanding performance for enterprise AI workloads, including training and real-time inference. Compared to B200 systems, they offer more performance per GPU, 50% greater GPU memory, and double the InfiniBand speed with 800G XDR non-blocking NVIDIA Quantum-X800 fabric.

​
Specifications

1 Usable storage is less than the raw capacity when configured as RAID. See RAID layout and throughput to learn about RAID configurations and their performance characteristics.

​
Primary use cases

Training foundation models from scratch (130B+ parameters), massive-scale inference, and developing frontier AI systems. DeepSeek V4 Pro, Kimi K2.6, GLM 5.1, and other future advanced GPT-5 class models, complex Mixture-of-Experts (MoE) models, and large multimodal systems built for coding and agentic workflows.
Last modified on June 23, 2026