Penguin Solutions Unveils Enhanced OriginAI Infrastructure with NVIDIA GPU Support

Scheduled for launch in early 2026, the latest iteration of Penguin's OriginAI will feature NVIDIA's cutting-edge DGX B300 and RTX PRO 6000 GPUs, designed for both training and inference workloads. The DGX B300, powered by the innovative Blackwell architecture, boasts an impressive training throughput of up to 72 petaFLOPs and an astounding 144 petaFLOPs for inference tasks. Meanwhile, the RTX PRO 6000 GPU, a dual-slot PCIe Gen 5 card equipped with 96 GB of GDDR7 memory, optimizes space and power consumption—key considerations for organizations looking to scale their AI deployments efficiently.

As an NVIDIA DGX-Ready Managed Services Provider, Penguin Solutions is committed to easing the challenges associated with AI deployment. By combining state-of-the-art GPUs with pre-validated infrastructure architectures and expert services, the company aims to accelerate AI adoption across various sectors. The pre-validation of both configurations ensures that organizations can expect faster, more reliable deployments with significantly improved time-to-value.

Dara Ambrose, vice president of products and solutions at Penguin Solutions, emphasized the importance of this integration, stating, “Integrating the latest NVIDIA GPUs into our OriginAI infrastructure solution is an important development in our ability to help organizations operationalize their AI investments.” This commitment to delivering high-performance solutions is designed to maintain a competitive edge for clients navigating dynamic technological environments.

The OriginAI infrastructure is crafted to provide robust support for demanding workloads, combining proven architectures with next-generation hardware and advanced cluster management software. Configurations can range from 1/4-pod setups to expansive deployments exceeding 24,000 GPUs. Penguin Solutions enhances this offering by incorporating leading storage solutions from industry titans such as DDN, Pure Storage, VAST Data, and VDURA.

To ensure optimal performance, the OriginAI solution utilizes Penguin’s in-factory burn-in and integration environment, which validates cluster performance and confirms production readiness prior to shipment. This meticulous process, paired with ICE ClusterWare software and expert managed services, positions Penguin Solutions to deliver sustained, peak performance for AI infrastructure.

A key component of the OriginAI solution is the ICE ClusterWare software. The latest update, announced at SC25 in St. Louis, MO, introduces multi-tenancy capabilities, allowing operators to efficiently manage complex, multi-user environments. Moreover, the software features anomaly detection and auto-remediation functionalities, which proactively address issues and identify potential performance degradations before they impact operations.

General availability of Penguin Solutions’ enhanced OriginAI solution, complete with the latest NVIDIA configurations and storage validation, is set for early 2026. With these advancements, the company reinforces its position as a driving force in the AI infrastructure landscape, helping organizations transform their AI ambitions into actionable results.

For more information about the OriginAI pre-validated infrastructures, interested parties can visit the Penguin Solutions website.

About Penguin Solutions

Penguin Solutions is dedicated to supporting customers in overcoming complex technological challenges across its computing, memory, and LED lines of business. By leveraging expert skills, experience, and strategic partnerships, the company empowers organizations to turn their technological aspirations into reality.