Super Micro Computer, Inc.
SMCI is still being valued like a low-quality hardware assembler even after guiding FY27 revenue to $65B-$72B, and that disconnect can support upside if backlog converts without another margin collapse.
SMCI is being priced as if its AI server demand is both temporary and low quality. That is too harsh at $43.26. The company just finished FY2026 at $39.1B of revenue, then guided FY2027 to $65B-$72B and disclosed $60B+ of new orders in Q4. That is not the profile of a business rolling over. At 11.85x earnings and 0.67x sales, the market is discounting either a sharp growth stall or a collapse in profitability. Those outcomes are possible, but current valuation already assumes a lot of failure.
The key to the long case is not the headline Q4 margin, because 17.6% gross margin was flattered by mix, tariff benefits, reserve changes, and contract timing. The key is whether SMCI can hold something near the Q1 guided 10.4%-10.8% gross margin while converting delayed Q4 shipments and scaling enterprise mix. If it can, then forward earnings power should improve materially even without another one-time boost. The company also has operating leverage still to come from higher rack throughput, with manufacturing capacity moving to 6,000+ racks/month and liquid-cooled capacity above 3,000 DLC racks monthly.
The second leg of the thesis is strategic relevance. AI infrastructure is shifting toward dense, power-hungry rack-scale systems where integration speed, cooling, and platform readiness matter more than in traditional commodity servers. SMCI is one of the few players built around that exact problem set. Management said enterprise/channel was 50% of Q4 sales, which matters because it suggests the opportunity is widening beyond a handful of hyperscalers. That broadening can support both durability and mix.
What keeps conviction low is simple. FY2026 EPS was only $3.65, down 5.6% despite 43.9% revenue growth, free cash flow was just $0.7B, debt is $6.7B, and the cash conversion cycle stretched to 149 days. This is still a business proving that explosive demand can translate into clean, repeatable cash earnings. The stock is worth owning because the upside from that proof is large, but it is not yet a set-and-forget compounder.
AI rack demand stays supply constrained, enterprise mix improves, and forward earnings expand enough for investors to pay materially more than 11.85x for a still-growing platform.
FY27 revenue ramps into management's range, margins normalize above FY2026 levels, and the stock closes part of its gap to peers without needing a full rerating.
Backlog converts slower than expected, Q1 margin guidance proves optimistic, and FY27 revenue lands near the low end with working capital pressure keeping earnings depressed.
No analyst coverage available to cross-check.
We don't buy stocks where they are, we buy them where they're going.
FY27 is the entire case. If SMCI can move from $39.1B in FY2026 revenue toward the midpoint of $68.5B in FY27 while keeping gross margin around the guided 10.4%-10.8% zone instead of falling back toward single digits, the current 11.85x P/E will look far too low for its forward earnings power. The moat today is speed, rack-scale integration, liquid cooling, and close GPU platform alignment, and that moat is real over the next 12 months but less durable over 3-5 years if OEMs and cloud customers internalize more design work. AI is still a net tailwind because demand is shifting from boxes to full racks and power-dense systems, where SMCI is advantaged, but the market will keep discounting the name until margins and cash conversion prove that this growth is economic, not just volumetric.
- FY27 revenue guidance of $65B-$72B versus FY2026 revenue of $39.1B implies another step-up in scale, even after 43.9% growth in FY2026.
- Valuation is compressed at 11.85x P/E and 0.67x P/S, far below peer averages of 29.8x and 7.3x, with the stock still 26.4% below its $58.78 52-week high.
- Q4 produced $60B+ of new orders and record backlog, while Q1 FY27 revenue guidance of $14.5B-$15.5B suggests delayed Q4 shipments are converting quickly.
- Demand is real, not theoretical: management disclosed $60B+ in new orders in Q4 and guided Q1 FY27 revenue to $14.5B-$15.5B, up sharply from Q4 revenue of $11.1B.
- The AI build cycle still favors SMCI's speed and integration model, with management expecting 80%+ AI-related revenue going forward and manufacturing capacity tracking to 6,000+ racks/month, including 3,000+ DLC racks/month.
- Execution improved materially in recent quarters: EPS beats were +40.8%, +36.1%, and +84.2% in the last three reports after a weaker 2025 patch, which supports higher confidence in near-term conversion of backlog.
- The setup is not crowded on price alone: shares are up 58.93% in 3 months and 36.1% above the 200-day average, but still sit 64.8% below the 5-year high of $122.90, leaving room if earnings power normalizes higher.
- Q4 profitability overstated the run-rate, as gross margin surged to 17.6% versus guidance of 8.2%-8.4% on favorable mix, tariff benefits, lower reserves, and contract timing, then immediately guided back to 10.4%-10.8% for Q1.
- FY2026 EPS fell 5.6% to $3.65 even as revenue grew 43.9% to $39.1B, showing that scale has not yet translated into stable per-share earnings.
- Working capital is stretched: cash conversion cycle worsened to 149 days from 106 days in Q3, a meaningful drag for a business doing only $0.7B of free cash flow on $39.1B of revenue.
- Short interest remains high at 16.89% of shares, reflecting real skepticism around durability of margins, conversion of backlog, and whether AI server demand remains concentrated in a few product cycles.
| Metric | Value | Context |
|---|---|---|
| Price target | $88.94 | Ground-truth target, 105.6% vs current $43.26, Low confidence |
| FY2026 revenue | $39.1B | 43.9% YoY growth |
| FY2026 EPS | $3.65 | -5.6% YoY, showing margin dilution despite revenue growth |
| Operating margin | 6.1% | Up from 5.7% in prior FY |
| Net margin | 5.3% | Still thin for current AI enthusiasm |
| Valuation | P/E 11.85, P/S 0.67, P/B 1.8 | All below peer references, with peer avg P/E 29.8x and P/S 7.3x |
| FY27 revenue guide | $65B-$72B | Major scale step implied versus FY2026 |
| Q1 FY27 guide | Revenue $14.5B-$15.5B, GAAP EPS $0.89-$0.98 | Points to strong sequential conversion of delayed shipments |
| Q4 FY26 gross margin | 17.6% | Far above 8.2%-8.4% guide, but included nonrecurring benefits |
| Leverage and liquidity | Debt/Equity 0.46, Debt/EBITDA 2.20, Current ratio 3.87 | Balance sheet is serviceable, but not pristine for a cyclical hardware name |
| Free cash flow | $0.7B | Modest versus $39.1B revenue, pressured by working capital |
| Price setup | -26.4% vs 52-week high, +36.1% vs 200-day average | Strong momentum, but not fully recovered |
The biggest risk is that FY27 revenue lands but margin does not, because Q1 gross margin is already guided down to 10.4%-10.8% from Q4's 17.6%, which would leave the stock looking cheap for a reason.
FY27 Q1 earnings on November 3, 2026, especially whether revenue clears the $14.5B-$15.5B guide and gross margin holds at or above the 10.4%-10.8% range.
Ownership is reasonably concentrated, with the top three institutions holding 43.2% and insiders owning 8.9%, which provides some alignment but also means the stock can move hard when institutional expectations change. Short interest is elevated at 16.89%, or 95.2M shares, though only 1.8 days-to-cover, so the setup can squeeze on strong numbers but the short base is not trapped by illiquidity. Near-dated options positioning is listed as positive gamma, which can damp some volatility until a catalyst breaks the range.
The company does not provide a detailed segment split here, but management said enterprise/channel reached 50% of Q4 sales, indicating a broader mix beyond hyperscale AI customers.
Backlog and product demand remain heavily tied to AI infrastructure, especially rack-scale systems and DLC deployments.
A higher enterprise/channel mix should help diversification and potentially support better margin stability than a pure hyperscale mix.
No geography breakdown was provided in the supplied data.
SMCI posted a major Q4 margin/EPS beat and guided FY27 revenue to $65B-$72B.
Management's record is mixed but improving. On the stale 2025 Q1 call, the company pointed to Blackwell timing, guided Q2 FY25 revenue to $5.5B-$6.1B, said gross margin would be down 100 bps sequentially, and highlighted only ~50% capacity utilization while waiting for new-chip supply. Since then, the business has scaled from $39.1B FY2026 revenue, expanded operating margin to 6.1% from 5.7%, lifted rack capacity to 6,000+ per month from 1,500+ DLC racks per month in Silicon Valley, and returned to a much better earnings cadence with the last three EPS beats at +40.8%, +36.1%, and +84.2%. The weak point is consistency: 2025 included several misses, FY2026 EPS still fell 5.6% despite strong revenue growth, and the latest quarter's huge margin beat clearly included nonrecurring help. Credibility has recovered from the 2025 uncertainty, but not enough to earn high conviction.
Looking to fiscal 2027, our momentum gives us strong confidence to target our revenue in the range of $65 billion to $72 billion as we are in the process of a historic infrastructure buildout.
Now turning to our outlook for Q1 fiscal year '27, we expect net sales to be in the range of $14.5 billion to $15.5 billion, GAAP diluted net income per share to be between $0.89 and $0.98, and non-GAAP diluted net income per share to be between $1.01 and $1.10.
Q4 non-GAAP gross margin was 17.6% versus our guidance of 8.2% to 8.4%. 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, and perhaps the subsequent quarter.
In our preannouncement, we disclosed over $60 billion in new orders, driving our order book and backlog to new record levels as we enter fiscal year 2027.
So looking forward, we will consistently grow our overall gross margin. Although we will still grow very aggressive, very faster than GPU, but we will focus much more than before on enterprise and CPU storage.
SMCI's edge is system-level speed. It packages GPUs, CPUs, networking, power, and liquid cooling into deployable AI racks faster than slower-moving OEMs, and that matters when customers are trying to stand up capacity around each new NVIDIA platform. The moat is meaningful today because integration, thermal design, and time-to-delivery matter, but it is only moderately durable because much of the core silicon roadmap sits outside SMCI's control and larger OEMs can narrow the gap over time.
Capital allocation has been functional, not shareholder-friendly. Free cash flow was only $0.7B, there were no buybacks and no dividends, and management instead raised $5.6B in equity/preferred financing to support capacity and working capital through the AI buildout. That is defensible given the growth opportunity, but it also tells investors that internal cash generation has not yet caught up with the revenue surge.
Rated BUY, Low conviction. The stock is cheap enough relative to the demand picture that upside can be large if FY27 lands anywhere near management's revenue plan and margins stay around the Q1 guide instead of slipping back toward prior troughs. This is not a quality-first story yet, because cash conversion and margin durability still need proof. What would change the view is simple: a miss on the $14.5B-$15.5B Q1 revenue guide, gross margin below 10.4%, or further deterioration from the already-stretched 149-day cash conversion cycle. If those do not happen, the discount valuation should narrow materially.
≣ Full research note tap to expand
Executive Summary
BUY, Low conviction. SMCI rates a BUY at Low conviction. The stock sits at just 11.85x earnings and 0.67x sales despite FY2026 revenue of $39.1B, FY27 revenue guidance of $65B-$72B, and a Q4 order intake of $60B+ that points to unusual demand visibility. The catch is that Q4's 17.6% gross margin was not clean, Q1 gross margin is guided back down to 10.4%-10.8%, and working capital has become the key risk with cash conversion cycle at 149 days.
Investment Thesis
SMCI is being priced as if its AI server demand is both temporary and low quality. That is too harsh at $43.26. The company just finished FY2026 at $39.1B of revenue, then guided FY2027 to $65B-$72B and disclosed $60B+ of new orders in Q4. That is not the profile of a business rolling over. At 11.85x earnings and 0.67x sales, the market is discounting either a sharp growth stall or a collapse in profitability. Those outcomes are possible, but current valuation already assumes a lot of failure.
The key to the long case is not the headline Q4 margin, because 17.6% gross margin was flattered by mix, tariff benefits, reserve changes, and contract timing. The key is whether SMCI can hold something near the Q1 guided 10.4%-10.8% gross margin while converting delayed Q4 shipments and scaling enterprise mix. If it can, then forward earnings power should improve materially even without another one-time boost. The company also has operating leverage still to come from higher rack throughput, with manufacturing capacity moving to 6,000+ racks/month and liquid-cooled capacity above 3,000 DLC racks monthly.
The second leg of the thesis is strategic relevance. AI infrastructure is shifting toward dense, power-hungry rack-scale systems where integration speed, cooling, and platform readiness matter more than in traditional commodity servers. SMCI is one of the few players built around that exact problem set. Management said enterprise/channel was 50% of Q4 sales, which matters because it suggests the opportunity is widening beyond a handful of hyperscalers. That broadening can support both durability and mix.
What keeps conviction low is simple. FY2026 EPS was only $3.65, down 5.6% despite 43.9% revenue growth, free cash flow was just $0.7B, debt is $6.7B, and the cash conversion cycle stretched to 149 days. This is still a business proving that explosive demand can translate into clean, repeatable cash earnings. The stock is worth owning because the upside from that proof is large, but it is not yet a set-and-forget compounder.
Key Metrics
| Metric | Value | Context |
|---|---|---|
| Price target | $88.94 | Ground-truth target, 105.6% vs current $43.26, Low confidence |
| FY2026 revenue | $39.1B | 43.9% YoY growth |
| FY2026 EPS | $3.65 | -5.6% YoY, showing margin dilution despite revenue growth |
| Operating margin | 6.1% | Up from 5.7% in prior FY |
| Net margin | 5.3% | Still thin for current AI enthusiasm |
| Valuation | P/E 11.85, P/S 0.67, P/B 1.8 | All below peer references, with peer avg P/E 29.8x and P/S 7.3x |
| FY27 revenue guide | $65B-$72B | Major scale step implied versus FY2026 |
| Q1 FY27 guide | Revenue $14.5B-$15.5B, GAAP EPS $0.89-$0.98 | Points to strong sequential conversion of delayed shipments |
| Q4 FY26 gross margin | 17.6% | Far above 8.2%-8.4% guide, but included nonrecurring benefits |
| Leverage and liquidity | Debt/Equity 0.46, Debt/EBITDA 2.20, Current ratio 3.87 | Balance sheet is serviceable, but not pristine for a cyclical hardware name |
| Free cash flow | $0.7B | Modest versus $39.1B revenue, pressured by working capital |
| Price setup | -26.4% vs 52-week high, +36.1% vs 200-day average | Strong momentum, but not fully recovered |
Financial Strength
The balance sheet is good enough, not a reason to own the stock by itself. Liquidity looks solid with a 3.87 current ratio, and leverage at 2.20x debt/EBITDA is manageable for a business with this growth profile, especially after the $5.6B equity/preferred raise. The weak point is cash generation quality: $0.7B of free cash flow on $39.1B of revenue is thin, and the cash conversion cycle jumping to 149 days says inventory and receivables are carrying the growth story. Margin trajectory improved at the operating line, but profitability still needs to become more repeatable before the market will pay up.
Competitive Position
SMCI's edge is system-level speed. It packages GPUs, CPUs, networking, power, and liquid cooling into deployable AI racks faster than slower-moving OEMs, and that matters when customers are trying to stand up capacity around each new NVIDIA platform. The moat is meaningful today because integration, thermal design, and time-to-delivery matter, but it is only moderately durable because much of the core silicon roadmap sits outside SMCI's control and larger OEMs can narrow the gap over time.
Management & Guidance
Guidance is bold and partly credible. Management guided Q1 FY27 revenue to $14.5B-$15.5B, GAAP EPS to $0.89-$0.98, non-GAAP EPS to $1.01-$1.10, gross margin to 10.4%-10.8%, and full-year revenue to $65B-$72B. The revenue guide deserves respect because Q4 included customer readiness delays in power, cooling, and networking, and management also cited $60B+ in new orders plus record backlog. The margin guide deserves more skepticism because Q4's 17.6% gross margin was explicitly helped by nonrecurring items and timing, so the next few quarters need to show that double-digit gross margin is sustainable without those tailwinds.
Segment Analysis
- Product revenue (Not disclosed): The company does not provide a detailed segment split here, but management said enterprise/channel reached 50% of Q4 sales, indicating a broader mix beyond hyperscale AI customers.
- AI-related revenue (80%+ going forward): Backlog and product demand remain heavily tied to AI infrastructure, especially rack-scale systems and DLC deployments.
- Enterprise and channel (50% of Q4 sales): A higher enterprise/channel mix should help diversification and potentially support better margin stability than a pure hyperscale mix.
- Geographic revenue (Not disclosed): No geography breakdown was provided in the supplied data.
Capital Allocation
Capital allocation has been functional, not shareholder-friendly. Free cash flow was only $0.7B, there were no buybacks and no dividends, and management instead raised $5.6B in equity/preferred financing to support capacity and working capital through the AI buildout. That is defensible given the growth opportunity, but it also tells investors that internal cash generation has not yet caught up with the revenue surge.
Management Execution & Track Record
Management's record is mixed but improving. On the stale 2025 Q1 call, the company pointed to Blackwell timing, guided Q2 FY25 revenue to $5.5B-$6.1B, said gross margin would be down 100 bps sequentially, and highlighted only ~50% capacity utilization while waiting for new-chip supply. Since then, the business has scaled from $39.1B FY2026 revenue, expanded operating margin to 6.1% from 5.7%, lifted rack capacity to 6,000+ per month from 1,500+ DLC racks per month in Silicon Valley, and returned to a much better earnings cadence with the last three EPS beats at +40.8%, +36.1%, and +84.2%. The weak point is consistency: 2025 included several misses, FY2026 EPS still fell 5.6% despite strong revenue growth, and the latest quarter's huge margin beat clearly included nonrecurring help. Credibility has recovered from the 2025 uncertainty, but not enough to earn high conviction.
Positioning & Flows
Ownership is reasonably concentrated, with the top three institutions holding 43.2% and insiders owning 8.9%, which provides some alignment but also means the stock can move hard when institutional expectations change. Short interest is elevated at 16.89%, or 95.2M shares, though only 1.8 days-to-cover, so the setup can squeeze on strong numbers but the short base is not trapped by illiquidity. Near-dated options positioning is listed as positive gamma, which can damp some volatility until a catalyst breaks the range.
Bull Case
- Demand is real, not theoretical: management disclosed $60B+ in new orders in Q4 and guided Q1 FY27 revenue to $14.5B-$15.5B, up sharply from Q4 revenue of $11.1B.
- The AI build cycle still favors SMCI's speed and integration model, with management expecting 80%+ AI-related revenue going forward and manufacturing capacity tracking to 6,000+ racks/month, including 3,000+ DLC racks/month.
- Execution improved materially in recent quarters: EPS beats were +40.8%, +36.1%, and +84.2% in the last three reports after a weaker 2025 patch, which supports higher confidence in near-term conversion of backlog.
- The setup is not crowded on price alone: shares are up 58.93% in 3 months and 36.1% above the 200-day average, but still sit 64.8% below the 5-year high of $122.90, leaving room if earnings power normalizes higher.
Bear Case
- Q4 profitability overstated the run-rate, as gross margin surged to 17.6% versus guidance of 8.2%-8.4% on favorable mix, tariff benefits, lower reserves, and contract timing, then immediately guided back to 10.4%-10.8% for Q1.
- FY2026 EPS fell 5.6% to $3.65 even as revenue grew 43.9% to $39.1B, showing that scale has not yet translated into stable per-share earnings.
- Working capital is stretched: cash conversion cycle worsened to 149 days from 106 days in Q3, a meaningful drag for a business doing only $0.7B of free cash flow on $39.1B of revenue.
- Short interest remains high at 16.89% of shares, reflecting real skepticism around durability of margins, conversion of backlog, and whether AI server demand remains concentrated in a few product cycles.
Valuation & Price Target
Engine price target $88.94 (+105.6% vs current), Low confidence.
- P/E Multiple: $105.07 (14% weight)
- PEG (growth-adjusted): $320.47 (1% weight)
- P/S Multiple: $493.73 (1% weight)
- Quality-adjusted: $45.6 (20% weight)
The rating stays aligned with the quant baseline BUY (Low conviction), but for a narrower reason than the model implies. The engine's $88.94 target is attractive, yet the peer framework is noisy because the average peer P/E of 29.8x comes with 4.6x dispersion and several peers are not clean comps for an AI server integrator with SMCI's volatility. The real support for a BUY is not the P/S Multiple $493.73 or PEG $320.47, which are clearly unusable here, but the combination of 11.85x earnings, FY27 revenue guidance of $65B-$72B, and recent operating beats against a still-depressed stock price of $43.26. No sell-side coverage is available, so there is no useful consensus anchor either way.
Risk Assessment
Risk score 35/100 (Low). The biggest risk is that FY27 revenue lands but margin does not, because Q1 gross margin is already guided down to 10.4%-10.8% from Q4's 17.6%, which would leave the stock looking cheap for a reason.
The Bottom Line
Rated BUY, Low conviction. The stock is cheap enough relative to the demand picture that upside can be large if FY27 lands anywhere near management's revenue plan and margins stay around the Q1 guide instead of slipping back toward prior troughs. This is not a quality-first story yet, because cash conversion and margin durability still need proof. What would change the view is simple: a miss on the $14.5B-$15.5B Q1 revenue guide, gross margin below 10.4%, or further deterioration from the already-stretched 149-day cash conversion cycle. If those do not happen, the discount valuation should narrow materially.
Go deeper in the app
Live charts, options analytics, the full earnings-call breakdown, price alerts, and every other stock — updated in real time.
Get Tickr