HPE Q3 FY2026 Revenue Jumps 33.7% as AI Infrastructure Demand Beats Guidance by 10%

Hewlett Packard Enterprise reported fiscal third-quarter 2026 results on September 2, 2026, with revenue up 33.7% year-over-year to $12.213 billion, beating the $11.892 billion analyst estimate. Adjusted earnings per share came in at $1.11, topping the $0.93 forecast by 18.4%, and next-quarter revenue guidance of $14.35 billion at the midpoint landed 10.2% above what analysts had been expecting.

The Margin Story Behind the Growth

Adjusted operating margin reached 12.8%, up 4.3 percentage points year-over-year — a substantial improvement that indicates the revenue growth is coming with genuine profitability gains, not being purchased through margin compression. That combination (strong growth plus meaningfully improving margins) is a more durable signal than either metric alone.

What's Driving It: AI Infrastructure Specifically

• Management attributed the results to an ongoing mix shift toward higher-value networking, cloud, and AI solutions within HPE's overall product portfolio
• AI infrastructure demand was cited as a key growth catalyst, with HPE specifically noting increasing customer investment in AI inferencing and agentic AI workloads — a distinct category from AI training infrastructure
• The scale of the guidance beat (10.2% above analyst expectations for the coming quarter) suggests management has visibility into sustained near-term demand, not just a favorable one-quarter comparison

What This Means for Enterprise Infrastructure Buyers

HPE specifically calling out AI inferencing and agentic AI as growth drivers — distinct from the AI training infrastructure that dominated earlier compute-deal headlines — is a useful signal that enterprise demand is broadening beyond training large models toward running them in production. For organizations planning their own AI infrastructure investments, this shift suggests inferencing and agentic workload capacity is becoming a more prominent line item across the industry, worth factoring into procurement planning alongside training-focused compute.