Sep 2, 2026 earnings call — Record quarter and raised outlook
HPE reported record Q3 revenue of $12.2 billion and raised fiscal 2026 and fiscal 2027 outlooks.
Revenue $12.2B (est $12.0B) · EPS $1.11 (est $0.93) · Next-quarter revenue outlook $13.9B–$14.8B
Key quotes
"AI has become a multiyear growth driver, expanding demand across our HPE portfolio."
"Order bookings and revenue for our networking products and services reached record levels despite supply constraints, which limited our ability to convert the higher demand into revenue in the quarter."
"Supply constraints continue to affect our ability to fulfill the increased customer demand."
Analyst Q&A 11
durability of AI demand
Q (Katherine Campagna · Goldman Sachs): Analyst asked if current demand is a sustained infrastructure cycle or just pulled-forward spend, and how much of the raised fiscal '27 outlook comes from the new hyperscale inference and Oracle deal versus the rest of the business.
A: Management said demand is exceptionally strong and that enterprise AI inferencing and many use cases are driving durable demand. They noted networking orders were record-breaking and that the Oracle and hyperscaler deals are included in parts of the networking and Cloud & AI growth guidance.
"Our guide is informed by what we see in the market, and the market is telling us the demand continues to be exceptionally strong." — Antonio Neri (CEO)
networking orders vs revenue
Q (Amit Daryanani · Evercore): Analyst asked why organic networking growth was only 10% while orders rose 36%, and requested more color on the Oracle networking deal.
A: Management said orders are growing much faster than revenue because supply availability is the limiting factor and they expect revenue to accelerate as supply improves. They described the Oracle deal as a multiyear, multi-gigawatt engagement using QFX switching (Broadcom Tomahawk 6), their software and AI Ops, and PTX routing with their Express 5 silicon for scale-out and scale-across.
"In fact, as we said in our prepared remarks, our orders are growing 3.5x faster than the revenue." — Antonio Neri (CEO)
enterprise server upgrade demand
Q (Aaron Rakers · Wells Fargo): Analyst asked about upgrade opportunity and whether unit growth will accelerate as AI moves into enterprise, and if HPE is changing strategy on pursuing hyperscalers.
A: Management said the hyperscaler customer is acting as an enterprise using HPE for multibillion-dollar AI inferencing for internal use, and that their strategy has not changed. They said units will depend on supply availability, and supply constraints have limited units so far but they expect units to strengthen as supply improves.
"This is about a multibillion-dollar AI inferencing for their own internal usage as an enterprise customer." — Antonio Neri (CEO)
Helios opportunity timing
Q (Joseph Cardoso · JPMorgan): Analyst asked why HPE did not include the AMD Helios opportunity in fiscal '27 outlook and what upside it could add to networking revenue and margins.
A: Management said they expect Helios infrastructure to be available for ordering later in the calendar year and they are building a large pipeline, but they wanted firmer schedules and deployments before including it in guidance. They emphasized the opportunity is large but not included in the 14% to 17% networking growth guide now.
"We felt that we wanted to see a little bit more as the schedule firm up and then eventually start deploying these capabilities." — Antonio Neri (CEO)
details on $3.5B inferencing deal
Q (Wamsi Mohan · BofA): Analyst asked what is included in the $3.5 billion inferencing deal, the economics, and whether HPE is changing its approach to hyperscale opportunities.
A: Management said they cannot name the customer but described it as a hyperscaler acting like an enterprise. They said they are selling traditional servers for AI inferencing for the customer's internal usage, not Tier 1 cloud infrastructure.
"We are selling traditional servers for AI inferencing that they will use for their own internal usage." — Antonio Neri (CEO)
supply constraints details
Q (Asiya Merchant · Citi): Analyst asked where supply constraints are most acute, timing for easing, and how long-term agreements are affecting supply.
A: Management said constraints persist in commodities like DDR5, DDR4 and NAND, and many parts are limited by wafer and clean room capacity which should improve some in 2027 but ultimately needs more wafer capacity. They said HPE has signed multiyear LTAs to lock capacity and is deciding how to use that capacity.
"My expectation personally after seeing the exceptional demand that we see in the market and talking with our suppliers, some of them we signed already multiyear LTA agreements to lock our capacity." — Antonio Neri (CEO)
server gross margin normalization
Q (Erik Woodring · Morgan Stanley): Analyst asked why Marie said traditional server margins would 'normalize' going forward despite strong demand and improving unit trajectory.
A: Management said Q3 benefited from revenue scale, disciplined pricing, programs like Catalyst and deal mix that boosted margins. They said going into Q4 and '27 the mix will shift to more AI revenue and the mix inside traditional servers will moderate, so margins will reflect that mix change rather than the Q3 confluence.
"As you get forward and you look forward into Q4 and then this does carry forward into the guide that we gave into '27, bear in mind that the mix of deals, specifically in AI will ship." — Marie Myers (CFO)
Oracle win as reference case
Q (Timothy Long · Barclays): Analyst asked whether the Oracle win can be used as a reference design for other hyperscalers or neo-clouds, and whether custom routing silicon mattered for the win and can drive further scale.
A: Management said yes to both: the win shows they have scalable products with AI capabilities and time-to-market leadership that can drive customer interest. They said their Express routing silicon is a strong differentiator for scale and that high-capacity routing is required as gigawatts of infrastructure are deployed.
"I mean, yes, because obviously, it proves on the first part of your question, that we have a scalable set of products that deliver the performance with the AI capabilities that now everybody is looking for to drive these self-driving kind of operations." — Antonio Neri (CEO)
supply confidence and FCF
Q (David Vogt · UBS): Analyst asked what improved on the supply chain versus 90 days ago to support the $4 billion revenue raise and how that gives confidence, and asked Marie about the drivers of improved free cash flow conversion.
A: Management said confidence comes from securing multiyear long-term agreements that lock capacity and let them plan usage, aligning demand and supply. They added cash flow improvement is driven by much less restructuring in '27 compared with '26 and lower working capital needs in networking due to faster inventory turn.
"So yes, the first part of your question, the answer is yes because obviously, we put numbers out there if we have the ability to fulfill it, and that came through the work our supply chain team have done to secure these multiyear long-term agreements that locks the capacity." — Antonio Neri (CEO)
server units vs ASPs
Q (Mark Newman · Bernstein): Analyst asked how much server growth was from higher ASPs versus unit growth, and whether supply constraints will ease so units accelerate further.
A: Management said they expect units to strengthen in Q4 as supply allows, noting 2026 unit growth was constrained so much of growth was ASP-driven. They said supply will remain a constraint and backlog will stay high, but their guide already reflects expected supply availability versus backlog and demand.
"Look, we expect in Q4 units to strengthen because of what we see in the market." — Antonio Neri (CEO)
customer hesitation and budgets
Q (Matthew Niknam · Truist): Analyst asked if customers show hesitation over pricing actions to offset higher memory costs and where incremental IT budgets for HPE products are coming from at large customers.
A: Management said they do not see hesitation now; customers initially were surprised by higher costs but concluded that moving faster on AI is necessary. They said budgets are increasing and customers prioritize replacing older infrastructure and investing in on-premise AI factories to gain cost, agility and compliance benefits.
"When you do it on-premise, and we do that ourselves and we share the number, we can see up to 60% cost benefits on a token basis." — Antonio Neri (CEO)