Supermicro reported a historically strong quarter, with revenue nearly doubling YoY and an unexpected jump in gross margin from an optimized product mix and deferred contracts. Management guided for robust but lumpy growth in FY27, driven by a record $60 billion order backlog from new AI infrastructure buildouts. The company also discussed a strategic shift towards higher-margin enterprise CPU and DCPBS solutions. Reported Q4 FY26 revenue of $11.1B, up 93% YoY, and an extremely high non-GAAP gross margin of 17.6% (up 750bps QoQ) driven by a favorable product mix and deferred contracts.
Supermicro reported a historically strong quarter, with revenue nearly doubling YoY and an unexpected jump in gross margin from an optimized product mix and deferred contracts. Management guided for robust but lumpy growth in FY27, driven by a record $60 billion order backlog from new AI infrastructure buildouts. The company also discussed a strategic shift towards higher-margin enterprise CPU and DCPBS solutions. Reported Q4 FY26 revenue of $11.1B, up 93% YoY, and an extremely high non-GAAP gross margin of 17.6% (up 750bps QoQ) driven by a favorable product mix and deferred contracts.
Order momentum is accelerating with over $60 billion in new orders during Q4, driving record backlog entering fiscal 2027. Management sees demand strengthening across enterprise, neocloud, and large data center customers, with nine customers over $1 billion in revenue in fiscal…
Reported gross margin was 17.47%, reinforcing the quarter's better-than-guided profitability.
Guidance · revenue to $14.75B
Guided FY27 revenue to $65-72B, suggesting a clear acceleration in growth, but maintained a wider-than-usual range to account for timing and customer readiness issues.
Management indicated AI demand is stronger than ever, with over $60 billion in new orders and AI solutions expected to represent greater than 80% of revenue going forward, including GPU-based and emerging CPU-based agentic AI workloads. They are positioning as a total AI data center solution provider (DCPBS) with liquid cooling and proactive service capabilities.
Order momentum is accelerating with over $60 billion in new orders during Q4, driving record backlog entering fiscal 2027. Management sees demand strengthening across enterprise, neocloud, and large data center customers, with nine customers over $1 billion in revenue in fiscal 2026 and a continued shift to AI-related solutions.
Management continues to expand manufacturing capacity globally, including a new 32-acre DCPBS campus in Silicon Valley, targeting over 6,000 racks per month total capacity. CapEx for Q1 fiscal 2027 is guided at $50-60 million, up from $28 million in Q4, with fiscal 2026 CapEx at $162 million.
Management expressed strong confidence in growth momentum, citing record orders and backlog, and emphasized a strategic shift to a higher-margin DCPBS total solution model.
Delays pushed revenue to the low end of guidance and could affect timing of future AI project ramps.
“Q4 revenue came in at $11.1 billion due to some short-term customer delay in power shortage, cooling, and Networking.”
Liquid cooling adoption is a key driver for AI data centers; delays in readiness impact revenue recognition.
“large data center always have power readiness, data center readiness. are concerned, especially liquid cooling.”
“Gross margins improved by 750 basis points sequentially due to a better than anticipated customer and product mix, including the deferral of several contracts from Q4 fiscal year 26 to Q1 fiscal year 27.”
“High volume GPU margin usually much lower. CPU, storage, IoT, enterprise application, on the other hand, have a higher margin. So we will try to balance between the two vertical.”
“Our AI solutions contributed approximately 60% of total revenue in Q4 versus over 80% in Q3 due to the timing of some large AI project ramps. Based on our backlog, we believe greater than 80% of revenues will be AI-related solutions going…”
“When we look at our backlog, we have improved terms. and which will help us on our cash flow conversion. So therefore, we expect that this will allow us to carry a greater volume of business.”
| Показатель | Период | Диапазон | Середина | Статус |
|---|---|---|---|---|
| Gross margin | FY2027 Q1 | 10.4%–10.8% | 10.6% | GUIDED |
| Revenue | FY2027 Q1 | $14B–$15.5B | $14.75B | GUIDED |
| Revenue | FY2027 | $65B–$72B | $68.5B | GUIDED |
| Дата прогноза | Показатель | Целевой период | Прогноз | Факт | Результат |
|---|---|---|---|---|---|
| FY2026 Q3 | EPS | FY2026 Q4 | $0.65–$0.79 | $1.70 | Met / beat |
| FY2026 Q3 | Gross margin | FY2026 Q4 | 8.2%–8.4% | 17.47% | Met / beat |
| FY2026 Q3 | Revenue | FY2026 Q4 | $11B–$12.5B | $11.1198B | Met / beat |
| FY2026 Q2 | Gross margin | FY2026 Q3 | 6.7% | 9.95% | Met / beat |
| FY2026 Q2 | Revenue | FY2026 Q3 | $12.3B | $10.243B | Missed |
| FY2026 Q1 | EPS | FY2026 Q2 | $0.46–$0.54 | $0.69 | Met / beat |
| FY2026 Q1 | Gross margin | FY2026 Q2 | 6.5% | 6.3% | Missed |
| FY2026 Q1 | Revenue | FY2026 Q2 | $10B–$11B | $12.6825B | Met / beat |
Confirms tight partnership and volume shipments across the full NVIDIA product line, positioning SMCI as a key early partner, but also emphasizes NVIDIA's own system-level design competition.
“Through our long-term NVIDIA partnership, we are shipping volume skew across the GB300, MVLR72, HGX B300, B200 MVLR4, and RTX 6000 Pro organized.”
… very focused on logistics and inventory management, significantly reducing inventory reserve and the expedite charge. Together, these operational disciplines will help moderate quarter-to-quarter margin fluctuation are driven by an even customer and product mix, supporting our goal of consistent growing gross margins. Turning to our key product domain, our system building block allows us to quickly optimize every major silicon platform. Through our long-term NVIDIA partnership, we are shipping volume skew across the GB300, MVLR72, HGX B300, B200 MVLR4, and RTX 6000 Pro organized. While preparing first to market Vora Rubin, we are MVLR72, Nubin HGX, and Vera C1, and other high-density Vera systems. With AMD, We launched complete new Helios product line and MI450 total solution alongside strong EPYC CPU, MI350 and MI355X momentum. Working with Intel, we brought Parcel Deck Edge AI system to market and shipping Xeon 6 Plus platform in volume. We also dedicated on developing product for the strong demand of ARM AGI process base, Conan Phoenix. Architecture optimized for high performance per one, influencing workloads. Demonstrating our silicon partners' deep confidence in our …
Shows expanding partnership with AMD beyond GPUs to include EPYC CPUs, strengthening SMCI's CPU and AI portfolio.
… supporting our goal of consistent growing gross margins. Turning to our key product domain, our system building block allows us to quickly optimize every major silicon platform. Through our long-term NVIDIA partnership, we are shipping volume skew across the GB300, MVLR72, HGX B300, B200 MVLR4, and RTX 6000 Pro organized. While preparing first to market Vora Rubin, we are MVLR72, Nubin HGX, and Vera C1, and other high-density Vera systems. With AMD, We launched complete new Helios product line and MI450 total solution alongside strong EPYC CPU, MI350 and MI355X momentum. Working with Intel, we brought Parcel Deck Edge AI system to market and shipping Xeon 6 Plus platform in volume. We also dedicated on developing product for the strong demand of ARM AGI process base, Conan Phoenix. Architecture optimized for high performance per one, influencing workloads. Demonstrating our silicon partners' deep confidence in our engineering excellence. To support a massive demand We continue to expand our physical footprint. In Silicon Valley, we recently announced our new 32-acre DCPBS campus, featuring advanced optical photonics networking lab and data center scale manufacturing, which …
Partnership with Intel continues in the edge AI and CPU server space, providing SMCI with a diversified product base.
… us to quickly optimize every major silicon platform. Through our long-term NVIDIA partnership, we are shipping volume skew across the GB300, MVLR72, HGX B300, B200 MVLR4, and RTX 6000 Pro organized. While preparing first to market Vora Rubin, we are MVLR72, Nubin HGX, and Vera C1, and other high-density Vera systems. With AMD, We launched complete new Helios product line and MI450 total solution alongside strong EPYC CPU, MI350 and MI355X momentum. Working with Intel, we brought Parcel Deck Edge AI system to market and shipping Xeon 6 Plus platform in volume. We also dedicated on developing product for the strong demand of ARM AGI process base, Conan Phoenix. Architecture optimized for high performance per one, influencing workloads. Demonstrating our silicon partners' deep confidence in our engineering excellence. To support a massive demand We continue to expand our physical footprint. In Silicon Valley, we recently announced our new 32-acre DCPBS campus, featuring advanced optical photonics networking lab and data center scale manufacturing, which brings our USA footprint to nearly 4 million square feet. Globally, our facilities in Taiwan, Malaysia, and the Netherlands are …
Highlights a new product line based on ARM architecture, expanding SMCI's addressable market and reducing dependence on x86 and NVIDIA GPUs.
… MVLR72, HGX B300, B200 MVLR4, and RTX 6000 Pro organized. While preparing first to market Vora Rubin, we are MVLR72, Nubin HGX, and Vera C1, and other high-density Vera systems. With AMD, We launched complete new Helios product line and MI450 total solution alongside strong EPYC CPU, MI350 and MI355X momentum. Working with Intel, we brought Parcel Deck Edge AI system to market and shipping Xeon 6 Plus platform in volume. We also dedicated on developing product for the strong demand of ARM AGI process base, Conan Phoenix. Architecture optimized for high performance per one, influencing workloads. Demonstrating our silicon partners' deep confidence in our engineering excellence. To support a massive demand We continue to expand our physical footprint. In Silicon Valley, we recently announced our new 32-acre DCPBS campus, featuring advanced optical photonics networking lab and data center scale manufacturing, which brings our USA footprint to nearly 4 million square feet. Globally, our facilities in Taiwan, Malaysia, and the Netherlands are also ramping strongly to meet the demand, putting our total manufacturing capacity on track to exceed 6,000 racks per month, including more …
The company is picking and choosing orders to improve profitability, balancing high-volume, low-margin GPU deals with higher-margin enterprise CPU and DCPBS business, potentially capping top-line growth. — This strategy of balancing margin over pure volume may slow SMCI's revenue growth but signals a shift towards a more profitable business mix, which could change competitive dynamics in the server market.
Yeah, guys. Thanks for taking the questions. I have two if I could. And congrats on the strong results and the ongoing improvement and profitability here. And let me start just with that. Charles, Dave, what's a good way to think about what fiscal year 27 gross margins can be. You benefited from mix in June. Sounds like you're absorbing some of that mix from deal push out in September. It's still a nice improvement, apples to apples and gross margin guide. Can you walk us through how we should think about sort of the puts and takes on the margins, mix, CPU, things like that, and number one, how should we think about what's the useful estimate for gross margin fiscal year 27 and what may be…
Okay, thank you for the question. Yes, I mean, we will be very careful control of balance between revenue and profitability. As you know, high volume GPU margin usually much lower. CPU, storage, IoT, enterprise application, on the other hand, have a higher margin. So we will try to balance between the two vertical. Especially last 12 months, we have continued to grow Salesforce in the enterprise and application, server application, storage. So looking forward, we will consistently growing our overall gross margin. Although we will still grow very aggressive, very fast in GPU, but we will focus much more than before on enterprise and CPU storage. And also the PCPPS product line is getting mature. So we are shipping more and more DCPBS Hardware and also software service and some switches as well. So DCPBS will be our long-term much better profit margin product line.