For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for Vicor Corporation (VICR) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-08-23): Designation Watch · Gem Score -5 (−100…+100 Quality+Value blend) · Quality 55 · Value -65 · Sentiment 35 (timing only, not weighted)
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
Vicor Corporation
VICR NASDAQVicor Corporation is a technology company that designs, manufactures, and markets high-performance modular power components and complete power systems for a wide range of electronic applications. The company focuses on solutions for efficient conversion and management of electrical power, offering products such as modular power converters, configurable power systems, power component modules, and related integrated circuits tailored to diverse system requirements. Its portfolio includes converters, filters, input modules, and custom power systems used in sectors that demand reliable, compact, and efficient power delivery, such as computing, communications, industrial, and automotive electronics. Vicor Corporation’s modular approach enables engineers to optimize power architectures for density, efficiency, and flexibility at the board and system level. Headquartered in Andover, Massachusetts, USA, Vicor Corporation serves original equipment manufacturers and system integrators globally, playing a specialized role in the electronic components industry by addressing complex power conversion needs in advanced electronic systems.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 2.61
Total Equity: $711.82M
Shares: 45,450,000
Total Debt: $0.00
Cash: $402.81M
EBITDA: $102.61M
Total Debt: $0.00
Cash: $402.81M
Revenue: $452.70M
Revenue: $452.70M
Revenue: $452.70M
Total Equity: $711.82M
Tax Rate: -25.4%
Equity: $711.82M
Total Debt: $0.00
Cash: $402.81M
Current Liabilities: $65.32M
Long-Term Debt: $0.00
Total Debt: $0.00
Total Equity: $711.82M
Shares: 45,450,000
Shares: 45,450,000
CapEx: -$20.32M
Shares: 45,450,000
Stock Price: $234.74
Net Income: $118.56M
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 15, 2026 11:39am (8d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | — | — | $405.1M | $359.1M | $452.7M |
| Cost of Revenue | $181.2M | $218.5M | $200.1M | $175.1M | $193.3M |
| Gross Profit | $178.2M | $180.6M | $204.9M | $184.0M | $259.4M |
| Operating Expenses | $122.6M | $153.4M | $153.6M | $185.3M | $177.6M |
| Operating Income | $55.6M | $27.2M | $51.4M | -$1.3M | $81.8M |
| Net Income | $56.6M | $25.4M | $53.6M | $6.1M | $118.6M |
| EBITDA | $67.3M | $41.0M | $68.6M | $17.3M | $102.6M |
| EPS | $1.30 | $0.58 | $1.21 | $0.14 | $2.63 |
| EPS (Diluted) | $1.26 | $0.57 | $1.19 | $0.14 | $2.61 |
Balance Sheet (Annual)
Last updated: Aug 15, 2026 11:19am (8d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Cash & Equivalents | $182.4M | $190.6M | $242.2M | $277.3M | $402.8M |
| Total Current Assets | $356.8M | $362.6M | $420.4M | $463.0M | $587.4M |
| Total Assets | $477.2M | $536.9M | $594.9M | $641.1M | $785.8M |
| Current Liabilities | $49.1M | $64.5M | $44.2M | $61.8M | $65.3M |
| Long-Term Debt | — | — | — | — | — |
| Total Liabilities | $53.3M | $72.6M | $53.8M | $70.8M | $74.0M |
| Total Equity | $423.9M | $464.3M | $541.1M | $570.3M | $711.8M |
| Retained Earnings | $217.6M | $243.1M | $296.7M | $302.8M | $421.4M |
Cash Flow (Annual)
Last updated: Aug 15, 2026 11:39am (8d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Operating Cash Flow | $54.4M | $22.9M | $74.5M | $50.8M | $139.5M |
| Capital Expenditure | -$47.8M | -$64.0M | -$33.5M | -$23.6M | -$20.3M |
| Free Cash Flow | $6.7M | -$41.0M | $41.1M | $27.2M | $119.2M |
| Acquisitions (net) | — | — | — | — | — |
| Net Debt Issued / (Repaid) | — | — | — | — | — |
| Dividends Paid | — | — | — | — | — |
| Stock Buybacks | — | $0 | $0 | $-497,000 | -$35.2M |
| Net Change in Cash | $20.7M | $8.2M | $51.6M | $35.1M | $125.5M |
Growth Trends (YoY %)
Last updated: Aug 15, 2026 11:39am (8d ago)| Metric | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|
| Revenue Growth | — | — | -11.4% | +26.1% |
| Gross Profit Growth | +1.3% | +13.5% | -10.2% | +41.0% |
| Operating Income Growth | -51.1% | +88.8% | -102.6% | +6,346.4% |
| Net Income Growth | -55.1% | +110.6% | -88.6% | +1,834.3% |
| EBITDA Growth | -39.1% | +67.4% | -74.8% | +492.6% |
Dividend History (Last 20)
Last updated: Aug 15, 2026 11:19am (8d ago)| Date | Dividend | Declaration | Record | Payment |
|---|---|---|---|---|
| 2011-08-05 | $0.15 | — | — | — |
| 2010-07-14 | $0.30 | — | — | — |
| 2008-08-21 | $0.15 | — | — | — |
| 2008-03-31 | $0.15 | — | — | — |
| 2007-08-10 | $0.15 | — | — | — |
| 2007-03-07 | $0.15 | — | — | — |
| 2006-07-13 | $0.15 | — | — | — |
| 2006-02-24 | $0.12 | — | — | — |
| 2005-08-09 | $0.12 | — | — | — |
Deep Analysis
Narrative Economics
market-narrative step).
AI Lens 4th lens · how AI reaches this business · 5-yr
2026-08-15AI accelerator racks have made current density and conversion loss the physical limit on compute, pushing power delivery from a commodity BOM line into a co-designed, high-value subsystem — exactly where Vicor's high-density modules, lateral/vertical point-of-load parts and patent estate sit, visible in gross margin moving 51.2%→57.3% and revenue +26% against a flat industry.
The architecture decision is made upstream: Nvidia/OCP reference designs and hyperscaler power teams specify the topology, and TI, Infineon, MPS, Delta and Flex can win those sockets with fab scale and price. Vicor can be designed out of the very trend that validates its thesis, or reduced to a second source with eroding ASPs.
Whether Vicor's modules and IP are embedded in the 800VDC / vertical-power-delivery generation of AI racks rather than merely adjacent to it. Observable in named design wins, capacity utilization of the Andover fab, and the size and duration of licensing/royalty income.
A patent estate around factorized power conversion and high-density packaging, in-house power-semiconductor and magnetics process control, and multi-year qualification history in systems where a converter failure kills a very expensive rack.
AI Lens thesis
AI reaches Vicor almost entirely through demand and physics, not through automation of its own work: the product is a physical, thermally and electrically constrained component, so cheap intelligence cannot substitute for it, cannot be routed around it by an agent, and cannot let a startup replicate its silicon, magnetics and qualification history. What AI does is raise the marginal value of watts saved and cubic millimeters recovered inside an accelerator rack, which is why mix shift alone lifted gross margin 6pp and turned a -0.4% operating margin year into 18.1% with $119M FCF. The risk is equally structural and non-AI in character: value capture. The specifying authority for AI power architecture sits with GPU vendors and hyperscalers, and the competing suppliers are far larger. So the AI lens verdict is a wide one — high exposure, favorable direction, unresolved magnitude, decided by socket ownership and licensing enforcement rather than by anything AI does to Vicor's cost structure.
What the market may be underestimating
Upside The patent-licensing path monetizes the architecture even in sockets Vicor loses — near-100% incremental margin, no capex, no capacity constraint — and the market tends to treat that income as one-off rather than as a structural claim on AI power conversion.
Downside If AI racks migrate conversion stages upstream into power shelves and sidecars supplied by Delta/Flex-class integrators, Vicor's dollar content per GPU can shrink even as AI capex booms — a mix loss disguised by unit growth, amplified by heavy concentration in a few accelerator programs.
Outcome range spread 55 · unresolved
Growth Outlook
Analyzed 2026-08-17 16:25The question every valuation on this page silently assumes: is this company likely to grow? Judged forward — the business, its category, the world — against what's already printed.
Claude Reading
First, my own read of the tape before touching the model outputs. The quarterly sequence is genuinely lumpy: rev goes $85.9M → $93.2M → $96.2M → $94.0M → $96.0M → $110.4M → $152.3M → $113.0M. That Q4 2025 $152.3M print is the entire narrative — strip it out and you have a company doing ~$100M/quarter with wobbling margins (2.7% in Q1'25, 42.9% in Q2'25, 18.3% in Q1'26). Net income margins swinging 2.7% → 42.9% → 25.6% → 30.6% → 18.3% in five consecutive quarters is not a "platform emerging" — it's licensing/royalty income or one-time IP settlements bleeding through (Vicor has litigation history against SynQor, Foxconn, etc.). The recent_earnings_yoy of 1834% and revenue CAGR of 5.7% are the tell: earnings are inflating on non-repeating items while the top-line is barely growing on a multi-year basis. Q1'26 stepping down to $113M from $152M is exactly what you'd expect if Q4 had a chunky IP/settlement bump.
The models are internally contradictory in a way nobody flagged. The rule-based classifier calls this "narrative_platform" and slaps warnings on every profit metric — but the Pre-Flight AI correctly identifies Vicor as a "mature, profitable power electronics manufacturer with 40+ years of operating history." These are incompatible. A 40-year-old component maker with a 5.7% revenue CAGR is not a narrative platform; it's a cyclical specialty semi trading at a narrative-platform multiple. The anomaly warnings telling me to discount P/E and ROE because "profit metrics are lagging indicators" have it backwards here — for VICR, profit metrics are the *leading* indicator that current earnings are non-recurring. The Synthesis and Market Forces layers correctly land on "priced for perfection," and the Thesis Evaluation's -7 score with bear mass slightly exceeding bull mass is the most honest read in the stack. The Market Narrative layer nails it: "unanchored, story-driven, 60% growth assumption not historically consistent." Believe that one.
A contrarian defending the price would argue: VICR's Gen5 ChiP and factorized power architecture genuinely solves the 48V-to-core-voltage problem for NVIDIA-class GPUs, and design wins have multi-year revenue tails once locked in. The $402.8M cash pile with zero debt, 57% gross margin, and $119M FCF give them runway to invest through cycles. If Q4'25's $152M is a real run-rate rather than a settlement, then 2026 revenue could print $600M+ and the multiple compresses via growth. Fair — but the Q1'26 drop to $113M is a fact, not a projection, and it argues Q4 was a spike, not a step-function. The insider activity reinforces this: eight separate sales on July 2, 2026 at these prices is not portfolio-rebalancing noise, it's insiders monetizing the narrative. No offsetting buys.
Valuation reality: at $10.82B market cap on TTM revenue of ~$472M (summing latest four quarters: 152.3+110.4+96.0+113.0), that's ~23x sales for a business with a five-year revenue CAGR under 6% and demonstrated $100M/quarter organic run-rate. Even generous scenarios — say $700M revenue in 2027 at 25% sustained net margin = $175M NI — put you at 62x forward earnings, still rich. A defensible fair value, applying 8x sales (still premium for specialty power semis; Monolithic Power trades ~15x but with cleaner growth) to a normalized $500M revenue base, lands around $85–110/share. To justify $234, you need the $152M quarter to be the new floor AND 25%+ revenue growth for three years AND margin retention against TI/Infineon/MPS. That's three independent bets. I agree with the Synthesis "High Conviction Required" framing but push harder: this is overvalued, not merely demanding conviction. The models collectively hedge; the numbers don't support the hedge.
GPT Reading
The raw numbers do not support a $10.8B equity value. Vicor is a genuinely high-quality component company: no debt, $402.8M of cash, 57.3% gross margin in 2025, $139.5M of operating cash flow, and $119.2M of free cash flow on just $452.7M of revenue. That is excellent economics for a hardware business. But the market is paying roughly 23.6x sales, 99x EV/EBITDA, and about 91x earnings for a company whose annual revenue was $405.1M in 2023, fell to $359.1M in 2024, then recovered to $452.7M in 2025. That is not a hypergrowth trajectory; it is a lumpy niche supplier with operating leverage. At the current price, investors are treating 2025 as the first year of a durable AI-driven step-function. The quarterly pattern says otherwise. Revenue went from $94.0M in Q1'25 to $96.0M in Q2, then $110.4M in Q3 and $152.3M in Q4, before dropping back to $113.0M in Q1'26. If this were the start of a smooth hyperscaler ramp, I would expect less violent seasonality and a stronger carry-through after the blowout December quarter.
What stands out most is the mismatch between revenue scale and valuation scale. Backing out the $402.8M of cash, enterprise value is still about $10.4B, or roughly 22.5x 2025 revenue and around 18x a simple annualized run-rate based on Q1'26 revenue of $113.0M. Even if I generously annualize Q1'26 net income of $20.7M, the stock is still around 130x that earnings power; if I use 2025’s exceptional $118.6M NI, it is still near 90x. To justify today’s price on fundamentals, Vicor likely needs to become not a $450M business but something like a $1.5B-$2.0B revenue company with sustained 25%+ operating margins. Nothing in the reported history proves that yet. The operating leverage is real — 2024 produced basically breakeven operating income on $359.1M of sales, while 2025 produced $81.8M of operating profit on $452.7M — but that same leverage cuts both ways. Q1'25 had only 2.7% net margin, Q2'25 had an eye-popping 42.9%, Q4'25 30.6%, then Q1'26 fell to 18.3%. Those swings look like mix, concentration, and shipment timing, not a business that deserves to be capitalized like a software platform.
The strongest bull case is that the market is correctly looking through current lumpiness because Vicor has a real architectural edge in power density and efficiency, and once designed into AI infrastructure it could ride multi-year platform adoption. There is data support for that argument. 2025 revenue grew 26% over 2024, gross profit rose from $184.0M to $259.4M, operating margin rebounded from negative 0.4% to 18.1%, and free cash flow reached $119.2M with only $20.3M of capex. Return metrics are also unusually strong for manufacturing, with ROIC of 33.2% and ROE of 16.7%, while the balance sheet gives them the ability to scale without financing risk. If Q4'25’s $152.3M revenue was not a one-off but the early evidence of major AI ramps, then the current multiple would compress fast on forward numbers. A company doing $600M-$700M soon at current gross margin structure would look much less crazy than trailing metrics suggest.
I still weigh the bear side more heavily because the burden of proof is extreme and the evidence remains too episodic. The annual revenue CAGR of 5.7% and the very recent history of a 2024 downturn matter more to me than one explosive quarter. The insider tape also does not help; the most recent cluster is all sales, and while small relative to market cap, it is inconsistent with a stock that is obviously mispriced low. More importantly, price already assumes the favorable scenario. At 23x sales, you do not need Vicor to be good; you need it to become category-defining. Yet the reported base business is still only around $100M-$115M per quarter outside one standout Q4, and that means any pause in AI capex, customer concentration issue, or competitive response could crush the multiple even if the company remains profitable and cash generative. This is a good business carrying an extraordinary story premium.
What would change my mind is simple: I need sustained scale, not one or two spikes. If the next 3-4 quarters show revenue holding above $140M per quarter, with gross margin staying above 55% and operating margin above 20%, then I would accept that Vicor has crossed into a structurally larger earnings regime and deserves a far richer multiple than traditional component peers. If instead revenue slips back toward the $100M-$115M range or margins retrace into low-teens territory, that would confirm 2025 was a concentrated upswing being extrapolated too aggressively. Until the company proves it can annualize something closer to $600M+ revenue with durable profitability, I think the stock is priced for an outcome that remains possible but far from demonstrated.
Grok Reading
The raw numbers show a genuine earnings inflection that is easy to understate if you only look at the multi-year CAGR. Vicor went from $359M revenue and near-zero operating income in 2024 to $453M and $82M of operating profit in 2025, with net income leaping to $119M and free cash flow of $119M on only $20M of capex. Gross margin sits at 57% and net margin at 26%; the balance sheet is fortress-like with $403M cash and zero debt. That is real operating leverage, not accounting noise. But the stock at $235 and a $10.8B enterprise is capitalizing that year at roughly 24× sales and 90× earnings. The long-run revenue CAGR is still only ~6%, and the most recent quarter already showed the lumpiness the bull case pretends is behind them: revenue fell from $152M in Q4 2025 to $113M in Q1 2026, crushing sequential momentum even as trailing margins remain elevated. Earnings power of ~$120M against an $11B market cap is a 1% free-cash-flow yield. That only works if the AI-power narrative delivers multi-year 25–30% top-line compounding from here; the historical base and the latest sequential drop do not yet underwrite it.
What stands out most is the gap between the profit model working at current scale and the valuation assuming Vicor becomes a multi-billion-dollar platform franchise. ROIC of 33% and ROE of 17% are excellent for a components maker, and the 26% YoY revenue print plus the Q2–Q4 2025 ramp prove demand can surge when design wins hit production. Yet the same data set shows 2023 revenue was already $405M—so 2025 is only a modest recovery-plus above the prior peak, not a structural re-rating of the addressable market. Insider activity is uniformly sales across the most recent window, which does not prove the thesis is wrong but is inconsistent with management treating $235 as a gift. The quantitative models flag “narrative_platform” and warn that profit multiples are lagging; that warning cuts both ways. The story is doing the heavy lifting, and at 22–24× revenue the story has already been paid for in full.
The strongest counter-argument is straightforward: margins and cash conversion have already proven the economic model, recent YoY growth is 26%, and if Vicor locks sole-source or near-sole-source positions in high-density AI power delivery, the current $450M revenue base can double or triple without proportional opex, making today’s multiple look ordinary in hindsight. A smart opponent would also note that FCF quality is clean, the company has no leverage, and the 5× price swing over the past year means the market itself is still discovering the right multiple—so $235 could be mid-cycle rather than peak. I weigh that differently because customer concentration risk and competitive convergence are not theoretical; they are the explicit bear mass in the thesis score, and a single hyperscaler delay or a TI/Infineon modular response would compress the multiple long before the income statement fully reflects it. Paying 90× for a business that just printed a down quarter after its best year is not “option value”; it is priced perfection.
I would flip to a constructive stance if two consecutive quarters re-accelerate above $140–150M with margins holding above 25% net, or if disclosed multi-year hyperscaler/auto design-win backlog visibly supports a path through $800M+ revenue inside three years without multiple compression. Until then the stock is a high-quality cyclical compounder wearing a platform multiple.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
Vicor runs a clean balance sheet: $402.8M liquid cash, zero net debt, Altman Z of 90.2, and 2025 FCF of $119.2M against a $10.8B market cap. Earnings integrity looks strong - OCF/NI of 2.55x, accruals -2.5% of assets, Beneish M at -2.19, and diluted share count essentially flat (0.3% CAGR since 2021). No sign of accounting stretch or dilution leakage; per-share economics are protected. 2025 was a step-function year: revenue rebounded from $359.1M in 2024 to $452.7M (+26%), gross margin expanded from 51.2% to 57.3%, operating margin swung from -0.4% to 18.1%, and net income went from $6.1M to $118.6M. That is real operating leverage in a high-performance power conversion franchise. The concern is that 2024 showed how fragile the earnings can be - operating margin actually went negative on a modest revenue decline, meaning the cost base is heavy and cyclicality bites hard. The founder-CEO Vinciarelli's steady drumbeat of small S-sales (95 sells / 0 buys, ~$31M) reads as programmatic distribution rather than a signal, but the complete absence of insider buying during a business inflection is notable.
Verify before trusting this (6)
- Customer concentration in 10-K (hyperscaler / AI accelerator exposure driving 2025 inflection)
- Whether Vinciarelli sales are under a 10b5-1 plan and remaining ownership stake
- Backlog and bookings trajectory to test durability of the 57.3% GM / 18.1% OM level
- Segment mix between legacy brick products and Power-on-Package / VPD platform
- IP litigation status (historical patent disputes with competitors) and any contingent liabilities
- R&D and capex intensity trajectory versus the new revenue base
Vicor is a debt-free, cash-generative power components business with a real 2025 inflection, but the price is doing heroic work. A ~$10.8B market cap on a company whose revenue base is a few hundred million and whose operating margin was negative just last year implies the market is capitalizing a 5x expansion into AI/data-center power delivery as if it were already locked in. The e2e synthesis flags exactly this: priced for 5x expansion into markets not yet captured. Earnings quality is genuinely high, so I do not haircut for accruals - but I do haircut heavily for cyclicality and the 2024 margin swing to negative, which tells me the through-cycle earnings power is well below the peak the market is extrapolating. Deserved value on a quality-adjusted, cycle-normalized basis is materially below spot. The bull case (Vicor as the power-plane standard) is plausible but is the base case in the price, not the upside. The bear case (modular power is not proprietary, TI/Infineon compete, cyclical semis) is the risk the price ignores. Fair verdict: rich, not catastrophically so, but with negative margin of safety.
Verify before trusting this (5)
- 2025 order book and design-win disclosures for AI/HPC power modules
- Gross and operating margin trajectory quarter-over-quarter and durability under a revenue pause
- Customer concentration - what % of growth is one hyperscaler
- Guidance vs consensus revenue growth assumptions embedded in the multiple
- Ongoing insider sale pace and 10b5-1 plan details
The active narrative here is a strong platform-monopoly story (modular power for AI data centers, EVs, edge) with moderate durability and a medium cult following. That archetype is exactly what a risk-on tape (+52, VIX 14) amplifies, and with beta 2.38 VICR is levered to any risk appetite in the market. Recent tape confirms it: +22% on record data-center demand in May, +24% on raised Q2 guide, and a 26% recent run vs a 5.7% long-term CAGR - the story is being paid for. Analyst tone and news flow are skewed to the bull side of the AI-power trade. Against that, the July 2 -19% air-pocket with no clean catalyst is a real warning: at story-driven multiples, sentiment is one datapoint away from a fast unwind, and the bear frame (60% growth priced in, cyclical semi exposure, no proprietary moat) is sitting there waiting. Macro rates (10y 4.63%, mkt PE 26.2) are a mild headwind to long-duration story stocks, but the risk-on regime is dominant for now. Net: the pressure still leans up, but it is a high-beta tailwind that can invert quickly.
Verify before trusting this (4)
- Next earnings/guide - any tone shift on data-center orders would violently re-rate sentiment either way
- Hyperscaler capex commentary from NVDA/AVGO/MSFT/META - the sector narrative that VICR rides
- Whether analyst targets are being revised up into the run or fading (divergence would flag exhaustion)
- Any repeat of the July gap-down pattern - a second unexplained air pocket would confirm sentiment is cracking
AI reaches Vicor almost entirely through demand and physics, not through automation of its own work: the product is a physical, thermally and electrically constrained component, so cheap intelligence cannot substitute for it, cannot be routed around it by an agent, and cannot let a startup replicate its silicon, magnetics and qualification history. What AI does is raise the marginal value of watts saved and cubic millimeters recovered inside an accelerator rack, which is why mix shift alone lifted gross margin 6pp and turned a -0.4% operating margin year into 18.1% with $119M FCF. The risk is equally structural and non-AI in character: value capture. The specifying authority for AI power architecture sits with GPU vendors and hyperscalers, and the competing suppliers are far larger. So the AI lens verdict is a wide one — high exposure, favorable direction, unresolved magnitude, decided by socket ownership and licensing enforcement rather than by anything AI does to Vicor's cost structure.
Verify before trusting this (8)
- Gross margin sustainability above 55%
- Power-limited deployment commentary
- Licensing royalty run-rate
- Rack power per accelerator trend
- AI data center capex commitments
- Non-AI industrial/defense demand recovery
- 800VDC topology reference designs
- Competitor point-of-load module launches
The world is re-architecting power, not just compute: as accelerator sockets pull hundreds of amps, power conversion has to move from the board edge to directly beneath the die, and that physical constraint is what pulled Vicor from a slow-growth modular-power niche into a spec-relevant position. That is a real, mechanism-backed change in the addressable opportunity. But two things temper it. First, the same constraint has mobilized every large analog vendor and the hyperscalers' own power teams, so the socket is contested rather than owned. Second, the earnings inflection here is disproportionately IP/licensing-flavored — highly profitable, but not the same asset as a compounding product franchise. Outside AI, the end-markets (EV, industrial, aerospace/defense) sit under a macro-headwind, high-long-rate backdrop that has historically produced Vicor's negative years. Net: the business direction is genuinely up and share is being taken inside a growing category, but the ceiling on durable compounding is well below what the current price arithmetic encodes.
Prediction unavailable. valuation-synthesis has no result for VICR — the prediction needs its fair-value anchors.