HPE, Nvidia Unveil New Solutions To Accelerate Time To Value For Generative, Agentic, Physical AI

Hewlett Packard Enterprise and Nvidia have announced new artificial intelligence solutions to accelerate time to value for enterprises deploying generative, agentic and physical AI. Nvidia AI Computing by HPE, the combined offering, will offer new enterprise AI solutions with new capabilities for a full-stack, turnkey private cloud for AI. These solutions will support enterprises in training, tuning, or inferencing their AI models.

"To fully harness the power of AI, HPE and Nvidia bring to market a comprehensive portfolio of AI solutions that accelerate the time to value for enterprises to enhance productivity and generate new revenue streams," said Antonio Neri, president and CEO of HPE.

Jensen Huang, founder and CEO of Nvidia, added, “Nvidia and HPE are delivering the full-stack infrastructure companies need to develop, train, and deploy AI at scale—from generative and agentic AI to robotics and digital twins. This collaboration accelerates the AI-driven transformation of business and unlocks new levels of productivity and innovation.”

HPE is expanding HPE Private Cloud AI, which offers a self-service cloud enabled by HPE GreenLake cloud, with support for the new Nvidia AI Data Platform. It will help transform data into intelligence by leveraging Nvidia’s computing, networking, AI software and enterprise storage.

HPE is adding a new developer system to its cloud AI portfolio, and will offer a unified edge-to-cloud data access through the HPE Data Fabric Software. Additionally, HPE Private Cloud AI now supports deployment of pre-validated Nvidia blueprints to accelerate time to value for AI workloads.

HPE will offer GPU optimisation capabilities that include observability of AI-native software stacks to manage performance of training and inference workloads running on Nvidia accelerated computing clusters. 

The offering enables IT operations to improve the way they manage, optimise and troubleshoot AI workloads across hybrid deployments.

Nvidia AI Computing by HPE will deliver the latest AI servers to support the full range of AI model training, fine-tuning, and inferencing with Nvidia Blackwell Ultra and Blackwell platforms. These servers can be deployed with Nvidia accelerated computing, networking and AI Enterprise software to improve performance, efficiency, and reliability.

These servers include:

• Nvidia GB300 NVL72 by HPE 

• HPE ProLiant Compute XD servers with new Nvidia HGX B300 platform 

• HPE ProLiant Compute DL384b Gen12 with Nvidia GB200 Grace Blackwell NVL4 Superchip 

• HPE ProLiant Compute DL380a Gen12 with new NVIDIA RTX PRO 6000 BlackwellServer Edition 

The latest HPE server portfolio will have full lifecycle security with a dedicated security processor called secure enclave. This will protect against firmware attacks and creates full line-of-sight throughout the supply chain. Additionally, through HPE’s iLO 7 technology, these servers will be the first with post-quantum cryptography that meets the requirements for FIPS 140-3 Level 3 certification, a high-level cryptographic security standard.