SAN JOSE, CA (GTC), Mar 18, 2026 – NVIDIA introduces NVIDIA BlueField-4 STX, a modular reference architecture for accelerated storage infrastructure. The platform enables enterprises, cloud, and AI providers to deploy systems that support long-context reasoning for agentic AI.
AI agents that operate across multiple steps and tools require faster data access than traditional storage systems provide. Traditional data centers offer high-capacity storage but lack the responsiveness needed for agentic AI workloads. As context grows, limits in storage and data paths can slow inference and reduce GPU utilization.
NVIDIA STX enables storage providers to build infrastructure that keeps data accessible for AI workloads. The first rack-scale system includes the NVIDIA CMX context memory storage platform, which extends GPU memory with a context layer for scalable inference and agentic systems, delivering up to 5x tokens per second compared with traditional storage.
“Agentic AI is redefining what software can do – and the computing infrastructure behind it must be reinvented to keep pace,” said Jensen Huang, founder and CEO of NVIDIA. “AI systems that reason across massive context and continuously learn require a new class of storage. NVIDIA STX reinvents the storage stack, providing a modular foundation for AI-native infrastructure that keeps AI factories operating at peak performance.”
NVIDIA STX integrates the NVIDIA Vera Rubin platform with a storage-optimized NVIDIA BlueField-4 processor. The design combines the NVIDIA Vera CPU, NVIDIA ConnectX-9 SuperNIC, together with NVIDIA Spectrum-X Ethernet networking, NVIDIA DOCA and NVIDIA AI Enterprise software.
The STX architecture provides up to 4x higher energy efficiency than CPU-based storage systems and can ingest 2x more pages per second for enterprise AI data. Storage providers co-designing AI infrastructure based on NVIDIA STX include Cloudian, DDN, Dell Technologies, Everpure, Hitachi Vantara, HPE, IBM, MinIO, NetApp, Nutanix, VAST Data, and WEKA.
AIC, Supermicro, and Quanta Cloud Technology (QCT) are building systems based on NVIDIA STX. AI labs and cloud service providers, including CoreWeave, Crusoe, IREN, Lambda, Mistral AI, Nebius, OCI, and Vultr, plan to use STX for context memory storage. STX-based platforms are expected to be available from partners in the second half of the year.
Source: NVIDIA
About NVIDIA
NVIDIA, founded in 1993 and headquartered in Santa Clara, CA, designs and manufactures graphics processing units, systems on chips, networking hardware, and AI intelligence software such as CUDA. Its products serve industries including gaming, data centers, autonomous vehicles, professional visualization, robotics, health care, and energy. The company introduced the GPU in 1999 and later expanded into accelerated computing and AI infrastructure. In gaming, its GPUs support high-performance rendering, while in AI and high-performance computing, its systems provide the infrastructure for training and deploying large-scale models. NVIDIA also develops tools for robotics and autonomous driving.