Отчётный звонок Super Micro Computer, Inc.
Record $13B+ in orders, including largest in history
SMCI reported a weak Q1 (revenue of $5B, below guidance) due to shipment delays but guided to a massive ramp in Q2 (>= $10.5B) and raised FY26 revenue to >= $36B. The company is pushing into a new strategic low-margin 'mega-scale' deal, likely for a neocloud customer, and is expanding capacity aggressively. Revenue of $5B missed guidance of $6-7B due to last-minute customer upgrades and logistics delays; $1.5B pushed to December quarter.
Вывод Buzzberg Record $13B+ in orders, including largest in history SMCI reported a weak Q1 (revenue of $5B, below guidance) due to shipment delays but guided to a massive ramp in Q2 (>= $10.5B) and raised FY26 revenue to >= $36B. The company is pushing into a new strategic low-margin 'mega-scale' deal, likely for a neocloud customer, and is expanding capacity aggressively. Revenue of $5B missed guidance of $6-7B due to last-minute customer upgrades and logistics delays; $1.5B pushed to December quarter. Читать полный анализСвернуть анализ
SMCI reported a weak Q1 (revenue of $5B, below guidance) due to shipment delays but guided to a massive ramp in Q2 (>= $10.5B) and raised FY26 revenue to >= $36B. The company is pushing into a new strategic low-margin 'mega-scale' deal, likely for a neocloud customer, and is expanding capacity aggressively. Revenue of $5B missed guidance of $6-7B due to last-minute customer upgrades and logistics delays; $1.5B pushed to December quarter.
- Record new orders of >$13B, primarily for NVIDIA GB300 platforms, including the largest single deal in company history.
- Gross margin guidance for December quarter implies ~6.5%, down 300 bps sequentially due to costs of ramping a new mega-scale platform.
- Management raised full-year revenue guidance to at least $36B, calling it 'very conservative'.
Что важно сейчас
Самые значимые изменения по итогам звонка.
Fiscal 2026 revenue guidance raised to at least $36B
DCPBS to drive higher margins, on track for data center solutions
Показать ещё 3 тезиса
Preparing for NVIDIA Vera Rubin and AMD Helios launches
Expanding capacity to support $100B+ potential revenue
Gross margins expected to trough in December quarter
Фактические результаты
| Показатель | Факт | Изменение |
|---|---|---|
| Выручка | $5.0178B | Факт |
| EPS | $0.35 | Факт |
| Валовая маржа | 9.31% | Факт |
| Операционная маржа | 3.63% | Факт |
| Свободный денежный поток | $-0.9498B | Факт |
| Капзатраты | $0.0323B | Факт |
Прогноз компании
| Показатель | Период | Диапазон | Середина | Статус |
|---|---|---|---|---|
| EPS | FY2026 Q2 | $0.46–$0.54 | $0.50 | Обозначен |
| Валовая маржа | FY2026 Q2 | 6.5% | 6.5% | Обозначен |
| Выручка | FY2026 | $36B | $36B | Повышен |
| Выручка | FY2026 Q2 | $10B–$11B | $10.5B | Обозначен |
Оценка менеджмента
Upbeat
Management expressed strong confidence in growth, highlighted record orders and revenue guidance raise, and emphasized their unique positioning in AI infrastructure, despite acknowledging margin pressures.
Оценка AI от менеджмента
Management highlighted strong demand for AI compute, with NVIDIA GB300 (Blackwell) products contributing over $13B in back orders, and expects continued growth from AI GPU platforms, including B300, B200, MI350/355X, and upcoming NVIDIA Vera Rubin and AMD Helios launches. They also see rapid growth in edge AI, agentic AI, and industry-specific models, and are positioning themselves as a leading AI
Инвестиции и мощности
Management is aggressively expanding global production capacity, with new facilities in Taiwan, Netherlands, Malaysia, and Middle East, targeting 6,000 racks per month (including 3,000 liquid-cooled racks) within the fiscal year. They are investing upfront in capacity, which pressures near-term margins, but expect cost efficiencies and economies of scale to improve profitability over time.
Компаниис момента звонка
Клиенты
Confirmed ongoing strategic partnership with xAI, indicating long-term customer relationship beyond the initial Colossus deployment.
Доказательства
“XAI, for sure, are very good partners. And whenever we have a chance, we try to work with them to learn something from them and to offer our best service. So, yes, XAI will continue to be our important partner for long term.”
Поставщики
SMCI received over $13 billion in new orders, largely for GB300-based racks, including its largest deal ever, indicating a significant demand spike for next-gen AI platforms from 'high-profile' customers. — The scale of these orders validates that the AI infrastructure buildout is in an early, accelerating phase and points to significant complexity and cost in the supply chain.
Доказательства
“We are shipping RTX Pro 6000, B200, MVL4, and AMD MI350, 355X platforms in volume to power generative AI, large language model, inference, and HPC workloads.”
SMCI's guidance implies ~6.5% gross margin for the December quarter, down from 9.5%, with management explicitly citing a 'strategic large design win' with 'higher costs and lower margin' as a major factor, likely referring to a hyperscaler or neocloud customer. — This indicates SMCI is taking a low-margin, high-volume strategic deal to secure market share, which signals intense price competition and immense pricing power among the leading AI chip customers.
Доказательства
“NVIDIA is getting available. we are receiving more and more allocation from them and preparing for a huge volume to ramp up, start from this quarter.”
Цепочка поставок
SMCI received over $13 billion in new orders, largely for GB300-based racks, including its largest deal ever, indicating a significant demand spike for next-gen AI platforms from 'high-profile' customers. — The scale of these orders validates that the AI infrastructure buildout is in an early, accelerating phase and points to significant complexity and cost in the supply chain.
Доказательства
“Our NVIDIA BlackWare Hotspot-based GB300 product line now at more than $13 billion in back orders, including the largest deal in our 32-year history.”
SMCI's guidance implies ~6.5% gross margin for the December quarter, down from 9.5%, with management explicitly citing a 'strategic large design win' with 'higher costs and lower margin' as a major factor, likely referring to a hyperscaler or neocloud customer. — This indicates SMCI is taking a low-margin, high-volume strategic deal to secure market share, which signals intense price competition and immense pricing power among the leading AI chip customers.
Доказательства
“...we are ramping a new product line at mega scale, and so, therefore, We're being a little conservative on the margin because we will have higher costs as we ramp production and shipment.”
Альфа цепочки поставок · 4с момента звонка
SMCI received over $13 billion in new orders, largely for GB300-based racks, including its largest deal ever, indicating a significant demand spike for next-gen AI platforms from 'high-profile' customers.
SMCI is ramping capacity to ~6,000 racks per month (including ~3,000 liquid cooled), suggesting a production capacity potentially exceeding $100 billion annually, indicating a major logistics and supply chain scale-up.
Доказательства
“If you time 3,000 liquid cooling tower per month, and you time 12 months a year, and each rack, for example, $3 million, so the number is more than $100 billion.”
SMCI's cash conversion cycle ballooned to 123 days with inventory at 105 days as it pre-builds for the December quarter, implying a cash-intensive ramp that could pressure suppliers' ability to keep pace.
Доказательства
“Q1 operating cash flow was impacted by lower net income and higher accounts receivable and higher inventory levels as we prepared for a strong Q2 with higher working capital needs.”
SMCI's guidance implies ~6.5% gross margin for the December quarter, down from 9.5%, with management explicitly citing a 'strategic large design win' with 'higher costs and lower margin' as a major factor, likely referring to a hyperscaler or neocloud customer.
Методология и полнота
Анализ основан только на заявлениях менеджмента. Показаны все подтверждённые упоминания компаний: 6. Фактические результаты и прогнозы разделены. Публичные доказательства ограничены восемью короткими атрибутированными цитатами. AI-анализ может быть неполным или ошибочным — проверяйте важные утверждения по первоисточнику.