Hewlett Packard Enterprise Expands AI Infrastructure with Multiple Partnerships

HPE used its HPE Discover Las Vegas 2026 conference to announce a series of AI infrastructure partnerships that significantly broaden its position in the enterprise AI market. The headline deal: Vultr, the world's largest privately held hyperscaler, selected HPE and NVDA NVIDIA to power its next-generation AI cloud infrastructure for cloud-scale data centers. Vultr will deploy the NVIDIA GB300 NVL72 systems by HPE, combined with NVIDIA Spectrum-X Ethernet networking, creating a production-ready platform optimized for high-performance model training and inference at enterprise scale.

The Vultr deal is part of a broader push at HPE Discover. HPE announced multiple industry collaborations aimed at expanding its AI factory ecosystem and accelerating agentic AI adoption across enterprise customers. The company also highlighted a 20% increase in Private Cloud AI token throughput achieved through closer engineering alignment with NVDA NVIDIA, a concrete performance benchmark that gives prospective customers a reason to choose the joint HPE-NVIDIA stack over competing infrastructure providers.

The flurry of partnerships reflects a strategic bet by HPE that the AI infrastructure build-out is still in early innings and that deep ecosystem alliances with chip and cloud partners can differentiate its offerings. With AI infrastructure growth rewriting data center design norms, HPE is positioning itself as the systems integrator of choice for enterprises that want validated, production-ready AI factories rather than assembling components from multiple vendors. Seventeen analysts revised earnings estimates upward following the Discover announcements, reflecting growing confidence in HPE's AI revenue trajectory.