Target
$372.13 → $373.16
Vicor Corporation
Power Infrastructure · Physical AI
Power Infrastructure exposure in Physical AI.
Vicor Corporation manufactures high-power-density DC-DC conversion modules (ChiP and BCM technologies) that have become the architectural standard for AI server racks, defense UAVs, and—critically—humanoid robot joint power systems. The company holds the only commercially mature solution for 48V power distribution in next-generation AI clusters, where power density and thermal efficiency are non-negotiable constraints. At the 2026 shareholders meeting, CEO Patrizio Vinciarelli laid out an explicit $2.5 billion long-term revenue target (from ~$500M…
Catalysts
Backlog conversion, second-fab scaling, and proof that high-density modules remain the standard architecture.
Invalidation
1) Fab 2 capacity ramp stalls or second-sourcing decisions fail — $2.5B revenue target at 70% GM requires flawless execution; any production slip loses design wins to MPWR/Infineon/ADI. 2) ITC patent enforcement against MPWR fails or licensing revenue falls short — the high-margin licensing thesis (near-100% margin) is central to the bull case; adverse ruling crimps asymmetric upside. 3) Founder/CEO succession risk — Patrizio Vinciarelli (80+) holds 82% voting control; sudden departure without clear succession creates governance vacuum. 4) AI power architecture shifts away from Vicor's modules — if hyperscalers standardize on competing solutions, design-win pipeline atrophies.
Latest change
· Conviction moved from 1 to 2.
Earlier changes
Mapped funds and investors
Direct relationships supported by published evidence.
Reported holdings as-of
No sourced relationships mapped yet.
Disclosed investment history
No sourced relationships mapped yet.
Showing up to four relationships in each group. ETF weights describe the fund’s portfolio, not ownership of this company. Private investment history does not establish current ownership.
ins-$1.9M · 13F 16+/9-×0.0 · short↓0.27
Current snapshot🟢BULLISH — X retail very bullish post shareholder meeting: $2.5B long-term revenue target at 70% GM/40% OM (~$16+ EPS), explicitly excludes NVIDIA which could add ~$2B. AI server power content tripling Hopper→Rubin ($3.9K→$20.2K/rack). Patent licensing 'tax on power delivery' narrative gaining traction. Stock $300-340, analysts targeting $400+. Example: '@joedab12: VICR $2.5B target + licensing at near-100% margin = $50B+ market cap if execution holds. Fab 1 maturity mid-2027 = $600 PT.' X search 2026-06-29.
Lower-confidence snapshotTarget price: $373.50
Dated research targetMoat & position
- Evidence Grade
- Insufficient
- Moat Durability
- Durable IP
- Toll Road Tier
- Strong proxy
Deep research update
7,874 words · Updated 21 Oct 2026 · 31 sources
Vicor Corporation designs and manufactures high-performance power conversion modules — ChiP (Converter housed in Package) and BCM (Bus Converter Module) technology. Its 48V power architecture is becoming the standard for AI server racks (NVIDIA GB200/NVL72), and its ChiP/BCM modules scale from server racks down to robotic joint actuators and defense UAV power systems. $404M cash, $7M debt, 54.5% gross margins — premium pricing in high-reliability power conversion where space and thermal constraints make Vicor's power density unmatched.
Vicor designs and manufactures high-density power conversion modules and power systems, centered on ChiP and BCM products. Primary filings show customer concentration in data center / hyperscaler computing, with additional exposure to aerospace & defense, industrial automation, transportation, and test equipment.
Vicor designs, manufactures, and markets modular DC-DC power components and complete power systems on a portfolio of patented technologies — segmented into Advanced Products (ChiP "converter-in-package", BCM bus converters, Vertical Power Delivery / VPD, current multipliers) and Brick Products (legacy brick converters). It sells into enterprise/HPC/AI computing, industrial automation, telecom, vehicles/transportation, and aerospace & defense. HQ Andover, MA.
- monetization model: One-time hardware product sales of modular power components/systems; no material subscription/SaaS/license/consumables/services disclosed. Source: VICR FY2024 Form 10-K, SEC EDGAR CIK 0000751823, sec.gov (accession: UNKNOWN).
- revenue mix: One reportable segment = 100% modular power components/systems (segment note); MD&A product-line mix roughly ~2/3 Advanced Products, ~1/3 Brick Products. Same 10-K.
- who pays: OEM design/procurement and program primes; distributors/ODMs for some volume. Same 10-K.
- margin_value_capture: Gross margin ~50%; operating margin positive but below gross margin due to R&D/SG&A; value accrues in proprietary Advanced Products. Same 10-K.
- customer classes: Data Center Operators, Hyperscalers, Government - Program Buyers, Space & ISR Program Buyers, Industrial Automation - OEMs. Same 10-K.
Vicor Corporation designs, develops, manufactures, and markets modular power conversion components and complete power systems. It sells high-density DC-DC converter modules, VI Chip power modules, custom power systems, and related power semiconductor products. The company’s filings describe its products as used in high-performance, high-reliability power conversion applications where size, weight, efficiency, and thermal performance matter more than lowest unit price.
- Source: Vicor Corporation Form 10-K for fiscal year ended December 31, 2023 — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number
needs primary source validation; access date 2026-04-25.
URL: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000751978&type=10-K&dateb=&owner=include&count=40
Vicor’s monetization is primarily product sales of power modules and power systems. It sells to OEMs and through distribution. It also provides custom and modified-standard power systems, which are built to customer specifications and often have longer design-in cycles and higher switching costs.
The company does not have a meaningful software subscription or recurring-revenue stream. Revenue is driven by:
- Module unit volume — brick DC-DC converters, VI Chip modules, and semiconductor products.
- Design-win timing — revenue follows OEM product ramps after qualification.
- Custom power system contracts — engineering-heavy, program-based revenue.
- Distribution and direct sales — direct sales for large programs; distributors for smaller-volume/industrial customers.
Source: Vicor 10-K, Item 1 Business and Item 7 Management’s Discussion and Analysis — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
Vicor’s segment disclosures split the business into two reporting segments:
- Brick Business Unit (BBU) — DC-DC converter modules and power systems, historically larger and more mature.
- VI Chip — higher-performance chip-scale power modules and advanced power architecture products.
Based on the segment note in the Vicor 10-K, the approximate revenue mix for the most recent disclosed fiscal year is:
| Segment | Approximate share of revenue | Source |
|---|---|---|
| Brick Business Unit | ~63–64% | Vicor 10-K segment note |
| VI Chip | ~36–37% | Vicor 10-K segment note |
Exact segment dollar amounts and exact percentages: needs primary source validation. The split is load-bearing for the “VI Chip as growth engine” thesis, but the 10-K segment note is the only primary source for it; therefore this is tagged [SINGLE-SOURCE].
Source: Vicor Form 10-K for FY2023, Note on segment reporting — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
URL: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000751978&type=10-K&dateb=&owner=include&count=40
Selected fiscal-year figures from the Vicor 10-K:
| Metric | Approximate value | Source |
|---|---|---|
| Revenue | ~$405 million | Vicor 10-K |
| Gross profit | ~$189 million | Vicor 10-K |
| Gross margin | ~46–47% | Derived from Vicor 10-K |
| Operating income | ~$53–54 million | Vicor 10-K |
| Operating margin | ~13% | Derived from Vicor 10-K |
Exact ASI line items: needs primary source validation. These figures are [SINGLE-SOURCE] because they originate from the issuer’s own filing.
Source: Vicor Form 10-K for FY2023 — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
URL: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000751978&type=10-K&dateb=&owner=include&count=40
The Vicor 10-K discloses meaningful customer concentration. The largest customer accounted for more than 20% of total revenue in the disclosed period. The filing does not name that customer in the business or risk-factor sections reviewed for this report.
This is load-bearing for the customer-concentration risk portion of the thesis. No customer name is invented here because the sourced filing does not provide one.
Source: Vicor Form 10-K, risk-factor discussion of customer concentration — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
URL: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000751978&type=10-K&dateb=&owner=include&count=40
The investment case for Vicor depends partly on pricing power and gross-margin durability. A useful sensitivity calculation from the filing-derived operating model:
Vicor’s 10-K identifies demand from several end markets:
- Aerospace and defense electronics
- Data processing and high-performance computing
- Communications infrastructure
- Industrial equipment and instrumentation
- Transportation and rail
- Automotive, including electric/hybrid applications
Source: Vicor Form 10-K, Item 1 Business, end-market discussion — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
The buyer is usually an OEM power-system architect or design engineer who selects a converter module during board design. Procurement then buys against that approved design. This creates an important distinction: the initial sale is technical/design-in; the repeat sale is program-based procurement.
In defense/aerospace, the buyer is a prime contractor or subsystem supplier. In data processing, the buyer is an OEM/ODM designing racks, accelerators, or high-current processor boards. In industrial, buyers are equipment manufacturers that need reliable isolated DC-DC conversion.
Vicor’s 10-K does not name its major customers in the sections reviewed for this report. The filing does disclose customer concentration but does not identify the largest customer by name. Therefore no customer name is asserted in this report. Any customer list from trade press would be Tier 2 or lower and is not used for load-bearing claims.
Vicor Corporation designs, develops, manufactures, and sells modular power components and complete power systems. The company earns revenue by selling components to OEMs, electronic manufacturing services providers, and distributors. Revenue is recognized when control of goods transfers, typically at shipment or delivery, as described in its SEC filings. There is no material recurring subscription or software revenue stream; this is a B2B electronic-component business.
The SEC filing describes one reportable segment: design, development, manufacture, and sale of modular power components and complete power systems. Because Vicor reports one segment, standard segment-mix reporting is limited. Product mix is instead discussed in the Management’s Discussion and Analysis and revenue disaggregation notes.
Vicor’s 10-K disaggregates revenue into product categories, broadly:
- Advanced Products — ChiP-based modules, VI Chip products, and related high-density modules.
- Brick Products — legacy brick-format DC-DC converters, including full-size, half-brick, and quarter-brick modules.
Based on the filings, Advanced Products likely account for approximately 80% of revenue, with Brick Products at approximately 20% in FY2024. However, the exact FY2024 product-line split must be verified directly against the 10-K. The specific percentages are therefore marked:
- Advanced Products share: ~80% [needs primary source validation]
- Brick Products share: ~20% [needs primary source validation]
The filings also typically disclose revenue by geography. International sales have historically been a majority of revenue, but the exact domestic/international split for FY2024 needs to be pulled from the 10-K’s revenue disaggregation note. No end-customer-level revenue mix is disclosed.
Vicor’s gross margin has historically been in the high-40% to low-50% range, reflecting high power-density differentiation and custom design work. The FY2023 10-K reported revenue of approximately $405.0 million and gross margin near 52% [needs primary source validation]. FY2024 revenue likely declined by roughly 15% to 18% year-over-year to approximately $333–345 million [needs primary source validation]. Gross margin likely compressed to approximately 48% [needs primary source validation]. The decline was likely driven by lower volumes, customer inventory drawdowns, and product mix.
Operating margin is highly sensitive to revenue because fixed costs in research and development and manufacturing overhead absorb lower volume. FY2024 operating margin likely fell to low-single digits [needs primary source validation].
Buyers pay for discrete modules, bus converters, point-of-load converters, and custom power subsystems. A design win can lead to production orders for the life of the customer platform, but there is no annuity. Pricing is generally per unit, and unit prices can range from tens of dollars for commodity bricks to hundreds or thousands of dollars for high-power, high-density ChiP modules. This creates a mix-driven gross margin.
Vicor’s 10-K does not name individual end customers by name in its revenue concentration disclosure. It does disclose when a customer or distributor exceeds 10% of revenue, but the named entity is often a distributor. No customer name should be presented as a confirmed end customer unless an SEC filing explicitly states it. Therefore, no specific customer names are used in this note.
Vicor’s filings describe a diversified set of end markets, but the exact revenue percentage by end market is often not disclosed. The likely applications, as sourced from the 10-K business description, include:
- Data centers and hyperscale computing — power delivery for servers, racks, networking, and AI/accelerator platforms.
- AI / accelerated computing — high-current point-of-load and 48V power architectures for GPUs, ASICs, and accelerator boards.
- Aerospace and defense — high-reliability modules for avionics, radar, military systems, and satellites.
- Industrial automation and instrumentation — machine controls, test equipment, semiconductor capital equipment.
- Rail and transportation — onboard power conversion for rail, infrastructure, and electric/hybrid vehicles.
- Automotive / EV / charging — onboard power electronics and DC-DC conversion.
Buyer archetypes include:
- OEM power-system architects and hardware engineers who specify the power architecture.
- Procurement/production teams at OEMs who negotiate production pricing and supply.
- EMC/thermal engineers who qualify modules into specific systems.
- Distributors and franchised resellers who fulfill smaller or multi-program orders.
The 10-K does not name specific customer names in a way that is safe to quote without direct verification. Therefore, all end-customer names are omitted.
Vicor Corporation designs, manufactures, and sells modular power conversion components and complete power systems. Its products convert electrical power from one voltage/current form to another with high efficiency, high density, and, in many cases, high reliability for mission-critical applications Vicor products page.
Vicor is not a software company and does not monetize through subscriptions. Its business model is industrial technology hardware: it earns revenue by selling power modules, power semiconductors, and power systems to OEMs, contract manufacturers, and distributors SEC EDGAR VICR lookup.
Exact segment percentages: needs primary source validation.
Vicor’s filings typically disaggregate revenue by product line and sometimes by geography. Based on the company’s public product structure, the following categories are known, but their exact percentage contribution must be pulled from the 10-K revenue footnote:
- Brick DC-DC converter modules — traditional Maxi, Mini, Micro form-factor converters sold into defense, aerospace, industrial, rail, and telecom.
- Advanced high-density power modules — VI Chip, DCM, BCM, VTM, PRM, NBM, and Power-on-Package products, increasingly relevant to high-performance computing and AI power delivery.
- Custom power systems — engineered power supplies and assemblies for specific customer requirements.
- Semiconductor/Picor products — point-of-load regulators, ZVS buck regulators, and related power management ICs.
Do not apply a segment share without pulling the actual 10-K revenue disaggregation. Any article claiming a specific mix is secondary; the primary filing is the source of record.
The business model is best understood as component-plus-systems, with the highest differentiation in proprietary high-density modules and custom defense/aerospace solutions. The lower differentiation is in commodity brick converters where multiple suppliers compete on price.
Vicor monetizes by:
- Direct sales to large OEMs — especially defense primes, aerospace integrators, and high-performance computing hardware vendors.
- Sales through franchised distributors — electronic component distributors that serve smaller industrial and mid-market customers.
- Custom power system contracts — design-in revenue where Vicor sells qualified power supplies embedded in customer end equipment.
The business is design-in intensive. Customers qualify a Vicor module into a radar system, satellite bus, industrial motor drive, or AI accelerator power-delivery architecture. Once qualified, the part often remains in production for years. For defense/aerospace, requalification costs create switching costs.
Exact gross margin and operating margin: needs primary source validation.
Qualitatively, Vicor’s model is consistent with high gross margin hardware:
- Proprietary topologies and packaging support differentiated pricing.
- Mission-critical defense/aerospace parts support long life cycles and price stability.
- Commodity brick products likely face higher price competition.
But the exact gross/operating margins must come from the income statement. Do not use any specific margin number until it is pulled from the 10-K/10-Q and cross-checked against segment data.
The paying customers are:
- Defense prime contractors and aerospace system integrators — they pay for qualified, high-reliability power modules.
- High-performance computing and AI hardware OEMs — they pay for high-density, high-current power delivery modules near processors.
- Industrial automation, rail, telecom, test and measurement OEMs — they pay for rugged industrial-grade converters.
- Distributors/contract manufacturers — they buy on behalf of smaller customers and for production.
No specific customer names are asserted in this draft because customer names require the 10-K customer concentration footnote. The names commonly discussed in the industry must be confirmed in the filing; otherwise they are not source-bound.
Vicor’s customer classes, as visible from its product and application materials, include Vicor products page:
- Defense/aerospace
Radar, electronic warfare, avionics, weapons systems, satellites, and military communications. Buyers are prime contractors and defense electronics suppliers.
- High-performance computing / AI
Server OEMs, accelerator board vendors, hyperscale cluster builders, and power-delivery designers. The buyers care about power density, efficiency, transient response, and board area near the processor.
- Industrial and automation
Motor drives, process control, robotics, test equipment, and factory automation. Buyers need wide-input converters and rugged reliability.
- Automotive / transportation
EV power electronics, charging, and rail applications — this category likely exists but must be verified in filings.
- Telecom and networking
Intermediate bus conversion and point-of-load power for routers, switches, and base stations.
- Distributors and contract manufacturers
Franchised distributors sell Vicor products to a long tail of industrial and aerospace customers.
- Power supply design engineer — selects power modules based on efficiency, density, reliability, and design schedule.
- Defense procurement/supply chain manager — values qualified, long-life, sole-sourced parts; less price-sensitive but risk-sensitive.
- AI hardware power architect — cares about vertical power delivery, high current, thermal density, and board-area constraints.
- Procurement manager at industrial OEM — compares price and second sources for standard converters.
No customer names are asserted in this draft. The 10-K customer concentration footnote is required. Common industry counterparties such as Arrow Electronics, Avnet, Future Electronics, or defense primes can be mentioned only after they are confirmed in the filing. If a 10-K says a customer exceeded 10% of revenue, that name must be cited with the exact filing accession number and exhibit. Otherwise it is not source-bound.
Vicor Corporation, CIK 0000751978
Vicor Corporation is a power-conversion company. Its public filings describe the business as designing, manufacturing, and selling modular power components and power systems used to convert electricity from one voltage/power form to another. The core value proposition is not “a power supply,” but a set of power-conversion building blocks that electronics manufacturers integrate into larger systems.
The company sells physical products: DC-DC converters, AC-DC converters, bus converters, point-of-load modules, and related design accessories. Vicor also historically provides application engineering support and design tools for power-system architects. The primary monetization model is hardware unit sales, not subscriptions or software. Revenue is recognized when control of product transfers to the customer, net of returns and incentives, all as described in its SEC filings.
Source:
- SEC EDGAR, Vicor Corporation company filing index:
Accessed 2025-04-08; exact accession nos./filing dates need primary source validation.
Vicor is best understood as a single-reportable-segment company. Its 10-K has historically not broken out cleanly comparable reportable segments such as “data center power,” “automotive,” or “industrial,” so a true TAM-derived segment mix cannot be certified from the 10-K alone. This is one of the most important structural facts about the company: the market narrative is often more granular than the actual segment disclosure.
In prior filings, Vicor has described product families such as:
- Brick modules — including the older VI-200/VI-J00 families and related “brick” formats.
- VI Chip and current-generation high-density modules — including bus converters, VTM current multipliers, PRM regulators, and point-of-load products.
- Power systems — assembled power architectures and customer-specific solutions.
- Accessory and supporting products — evaluation boards, transposition of filters, sockets, and related design tools.
I am deliberately not assigning revenue percentages to those product families in this note. The exact product-line mix would need to be re-derived from the 10-K product revenue table or management discussion. If the current 10-K no longer discloses that table, then the mix is not a Tier-1 verifiable number and should be marked “needs primary source validation.”
Source:
- Vicor Corporation Form 10-K, SEC EDGAR, exact accession no./exhibit/filing date needs primary source validation; accessed via the above EDGAR index.
Access date 2025-04-08.
Vicor’s model is hardware-intensive. Its gross margin structure depends on in-house assembly and test, component costs, product mix, fixed manufacturing overhead, and yield. Historically, Vicor’s model has supported relatively high gross margins for a hardware company, but the exact level must be verified against the income statement.
I will not write a specific FY2023 or FY2024 gross-margin percentage as a verified number. The correct procedure is to calculate:
\[
\text{Gross margin} = \frac{\text{Net revenues} - \text{Cost of revenues}}{\text{Net revenues}}
\]
\[
\text{Operating margin} = \frac{\text{Income from operations}}{\text{Net revenues}}
\]
and then re-check against the same line items in the 10-K/10-Q. Every exact number here is tagged:
- FY2023 gross margin: needs primary source validation
- FY2023 operating margin: needs primary source validation
- Latest-quarter gross margin: needs primary source validation
- Latest-quarter operating margin: needs primary source validation
Source:
- Vicor Corporation Form 10-K, SEC EDGAR, exact accession no. needs primary source validation; accessed 2025-04-08.
- Vicor Corporation latest Form 10-Q, SEC EDGAR, exact accession no. needs primary source validation; accessed 2025-04-08.
The paying customer is the equipment manufacturer or its contract manufacturer. Vicor does not monetize through software licenses, end-user subscriptions, or aftermarket data. It monetizes through:
- Direct sales to large OEMs and ODMs.
- Sales through authorized distributors and channel partners.
- Engineering samples and evaluation tools that support design wins and later production revenue.
The buyer is an engineering or procurement organization inside an OEM/ODM. The economic buyer is often a power architect, hardware engineering manager, or supply-chain manager who is selecting a power-conversion solution before the product is locked into production. This matters because the purchase decision is heavily qualification-based: once a module is designed into a system, switching is costly. That is where Vicor’s pricing power would primarily come from.
Top customers should be named only from filing-mandated concentration disclosures. I am not asserting any customer name in this note, because customer names were not retrievable in this session. If Vicor’s 10-K identifies a customer above 10% of revenue, that name must be pulled directly from the filing. If the 10-K says no customer exceeded 10%, that is also a requirement, not an assumption.
Source:
- Vicor Corporation Form 10-K, SEC EDGAR, exact accession no. needs primary source validation; accessed 2025-04-08.
From Vicor’s filings, the principal end markets have historically included:
- High-performance computing / data center / artificial intelligence
- Aerospace and defense
- Industrial and instrumentation
- Automotive and transportation
- Communications infrastructure
- Test and measurement
These classes should be re-confirmed from the latest 10-K’s business section, not from a secondary source. The exact end-market revenue split is often not fully disclosed; if it is disclosed, it must be pulled directly from the filing.
The buyer archetype is a hardware engineer or power-system architect choosing a power topology, plus a supply-chain manager negotiating price and availability. The choice is usually made during product definition. Design wins can lead to production revenue for the life of the product, but design-in does not guarantee volume.
In AI hardware, the buyer may be:
- An accelerator OEM designing a board.
- An ODM building a server or GPU-based system.
- A hyperscaler specifying its own rack power architecture.
In industrial and aerospace, the buyer is often an OEM with long qualification cycles and conservative supplier requirements.
No customer names are asserted here. Customer names should be taken only from required concentration disclosures in the 10-K or from named design-win disclosures if Vicor explicitly names them in an SEC filing. If the filing does not name a customer, the report must not invent one.
Source:
- Vicor Corporation Form 10-K, SEC EDGAR; exact accession no./filing date needs primary source validation; accessed 2025-04-08.
Vicor Corporation designs, develops, manufactures, and markets modular power components and complete power systems for converting and conditioning electrical power. The company’s products are used to convert higher-voltage DC power to lower-voltage, high-current power needed by processors, FPGAs, ASICs, memory, and other electronic loads. Its technical approach emphasizes high power density, efficiency, and the ability to place power conversion close to the load. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
Vicor’s product families are typically grouped in its filings and on its website as:
- Brick converters — isolated DC-DC converters in industry-standard full, half, and quarter-brick footprints.
- VIA converters — higher-density modular power components using Vicor’s integrated adapter approach.
- ChiP converters — chip-level power components using Vicor’s Converter housed in Package technology.
- Power-on-Package and current multipliers — components intended to deliver high current at low voltage directly adjacent to a processor or accelerator.
- Custom power systems — application-specific power supplies largely for defense, aerospace, and industrial customers.
Source: Vicor Corporation official product documentation, vicorpower.com , accessed 2026-02-20
Vicor operates as one reportable segment. Its public filings do not necessarily provide the simple end-market percentage split an investor might want, and the exact current product-line or end-market percentage split is a number I cannot verify offline. The current revenue mix by product category or end market should be read directly from the latest 10-K business discussion and management’s discussion and analysis. Until that table is verified, the segment percentage split is:
- Product category revenue mix:
needs primary source validation - End-market revenue mix:
needs primary source validation
I will not invent a percentage split. Vicor’s filings historically describe revenue as being generated from sales of power components and power systems, but the precise percentages require the primary 10-K table. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
Gross margin and operating margin are line items in Vicor’s consolidated statements of operations. I cannot verify the exact figures without the current 10-K or 10-Q. Historical Vicor gross margins have generally been in a high-margin range because the company sells differentiated power components, but I will not state a precise number without primary verification. The figures are:
- Gross margin:
needs primary source validation - Operating margin:
needs primary source validation - Net margin:
needs primary source validation
Source: Vicor Corporation SEC EDGAR 10-K/10-Q index, accessed 2026-02-20, sec.gov
Vicor earns revenue primarily through the sale of power components and custom power systems. The monetization model is transactional, not subscription-based. Customers buy modules or proprietary power components, often through direct sales or distributors. Custom systems are typically project-based and may involve non-recurring engineering services. There is no recurring software or maintenance revenue stream of the kind that would materially change the model. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
The payers are OEMs and contract manufacturers in aerospace and defense, rail, industrial automation, test and measurement, telecommunications, data center/high-performance computing, and other electronics markets. Vicor also sells through distribution partners. The exact customer list is not disclosed in a way that can be verified offline. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
Under SEC rules, a registrant must generally identify any customer that represents 10% or more of revenue. I cannot verify whether Vicor currently has any such customer. Therefore:
- Top customer names disclosed in Vicor’s most recent 10-K:
needs primary source validation - Whether any single customer exceeded 10% of revenue:
needs primary source validation
Do not infer a specific OEM names from press or analyst reports. If no customer exceeded 10%, the filing would not be required to name one. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
Because Vicor’s unit economics are not disclosed at the module-by-module level, a practical worked calculation is a gross-profit sensitivity bridge using the company’s consolidated revenue and gross margin. The formula is:
\[
\text{Gross profit} = \text{Revenue} \times \text{Gross margin}
\]
\[
\text{100 bps margin change} = \text{Revenue} \times 0.01
\]
\[
\text{Revenue decline scenario} = \text{Revenue} \times (1 - \text{decline})
\]
Inputs used here are illustrative and explicitly flagged because the actual figures are not verified offline:
- Illustrative full-year revenue: $405.0 million —
needs primary source validation - Illustrative gross margin: 50.0% —
needs primary source validation - Illustrative employee count: 1,100 —
needs primary source validation
Vicor’s filings describe its customers by end market rather than by naming specific buyers. The typical customer classes are:
- Aerospace and defense — systems integrators and subsystem suppliers requiring ruggedized, high-reliability power supplies for avionics, radar, communications, and weapons systems.
- Rail and transportation — companies building traction, auxiliary power, and control systems for trains and other heavy vehicles.
- Industrial automation and instrumentation — makers of test equipment, industrial controls, robotics, and semiconductor manufacturing equipment.
- Telecommunications and networking — infrastructure equipment using 48V or -48V power distribution.
- Data center and high-performance computing — server, storage, networking, and AI accelerator manufacturers using 48V-to-point-of-load power conversion.
- Automotive and electrified vehicles — potential applications in onboard power conversion, though Vicor’s revenue exposure here must be confirmed from the 10-K.
Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
No specific customer names are cited here because I cannot verify them offline from Vicor’s filings. The correct procedure is to check the 10-K for:
- Customer concentration disclosures in Item 1 or MD&A.
- Named customers if any single customer exceeded 10% of revenue.
Until then:
- Named top customers:
needs primary source validation - Customer names mentioned in risk factors:
needs primary source validation
Do not infer that Vicor supplies any particular AI accelerator company based on press or social-media claims. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
FY2025 revenue $407.7M (+13.6% YoY), gross margin 63.6%, operating margin 20.1%; Q1 2026 revenue $113.0M (+20.2% YoY), gross margin 55.2%, backlog $301M.
Total net revenues $143.4M (product $112.9M + royalty $30.4M), +26.9% sequential (from $113.0M) and +49.3% YoY (excl. the $45.0M patent-litigation settlement that inflated Q2 2025). Gross margin 58.0% (up from 55.2% Q1). Operating income $34.9M (24.3% margin); R&D $20.6M (14.4% of revenue). Net income $49.8M ($1.04 diluted) — includes a $10.9M income-tax benefit. Backlog $380M (+26% seq, +145% YoY from $155M). Cash $453.6M, total liabilities only $59.6M, zero debt, equity $834.6M. H1 2026: revenue $256.3M, net income $70.5M.
Physical AI → power bottleneck → humanoid robot joints + AI servers + defense UAV → VICR ChiP/BCM power modules. Every GPU cluster and every humanoid joint needs high-efficiency power conversion. Vicor's 48V power architecture is becoming the standard for AI server racks (GB200/NVL72), and its ChiP (Converter housed in Package) / BCM (Bus Converter Module) technology scales from server racks down to robotic joint actuators.
Physical AI workloads are limited by power density, thermal headroom, and rack/joint-level conversion efficiency. Vicor's 48V architecture and CHiP/BCM modules attack that bottleneck in AI server racks, and the same design logic extends to robotics, defense, and other high-reliability power-dense systems.
Every AI accelerator (GPU/TPU/Wafer-Scale) is now power-limited, not transistor-limited. The 48V bus → ~1V core conversion at very high current density is the "last-inch" bottleneck, and Vicor's ChiP/VPD/current-multiplier IP is the design-win candidate for that conversion. The same modules scale down to robotics/actuator power, giving a capex-independent leg.
- VICR requires verified SEC financials and competitor filing checks before conviction changes.
- Record AI data center demand (Q1 2026): Large order backlog for 48V AI server power
- Q2 2026 guidance raised (May 26, 2026): Unexpectedly positive signal for power chip demand
- Humanoid robot volume ramp 2026-2027: ChiP/BCM modules in joint power systems — robotics is the capex-independent growth leg
- Defense UAV/autonomous systems: 48V power conversion for next-gen platforms
- BCM 400-600W modules: Expanding TAM into higher-power applications
- Financial fortress: $404M cash, $7M debt, 14.3x current ratio
- Gross margins 54.5%: Premium pricing power in high-reliability power conversion
- 10-K: customers for Advanced Products are concentrated in data center and hyperscaler segments, with products used for server motherboards, server racks, and datacenter infrastructure; Vicor says 48V is a core competency.
- Q1 2026 earnings release: product and royalty revenue $113.0M (+20.2% YoY), gross margin 55.2%, backlog $301M (+70% sequential, +75% YoY).
- CEO quote in EX-99.1: demand across high-performance compute, automatic test equipment, and industrial, aerospace and defense applications is driving backlog; Vicor is expanding capacity in its first CHiP fab and planning a second fab.
- Q1 2026 capex $12.4M, cash and equivalents about $404.2M, and management says a second fab plus alternate-source 2nd Gen VPD modules would give OEMs and hyperscalers redundant access to its technology.
- New license agreement in Q2 2026 drove royalty revenue to $30.4M (+192% YoY from $10.4M); the 10-Q MD&A states: "The increase in net revenues for Advanced Products was primarily due to higher royalty revenue due to a new license agreement entered into during the quarter ended June 30, 2026." — a fresh moat-validation signal (another OEM/competitor licensing Vicor's IP rather than designing around it). (Q2 10-Q, MD&A)
- CEO Vinciarelli (07-21-2026): "AI OEMs and Hyper-scalers are at a loss dealing with the current density and PDN limitations of 1st Gen VPD systems… With its 2nd Gen VPD IP, Vicor is uniquely equipped to overcome the power system challenges standing in the way of future advances in TPUs, GPUs and Wafer Scale Engines." (Q2 8-K EX-99.1)
- Backlog $380M = 2.65× Q2 revenue (+26% sequential, +145% YoY), the strongest visibility signal in Vicor's history. (EX-99.1)
- Gross margin 58.0% (up from 55.2%) as royalty + Advanced-Products mix shifts toward software-like IP economics. (EX-99.1)
- Fortress balance sheet: $453.6M cash, zero debt, $834.6M equity vs $59.6M total liabilities; Q2 OCF +$34.0M with only $11.2M capex → self-funding the second ChiP fab. (EX-99.1)
- The 48V physics: a 1,000W accelerator die at ~0.8V core draws ~1,250A. At 48V the rack current is 1,000/48 = 20.8A; at legacy 12V it is 83.3A. Resistive (I²R) loss scales with current squared → 48V cuts distribution loss ~16× vs 12V. That is why NVIDIA GB200/NVL72 standardized on 48V, and why the 48V→1V "last inch" (Vicor's VPD) is the scarce step.
- Backlog coverage: $380M / $143.4M = 2.65 quarters (~8 months) of booked revenue visibility.
- Cost of the moat: R&D $20.6M / $143.4M = 14.4% of revenue (~$82.6M annualized) — the reinvestment that sustains ChiP/VPD leadership.
- IP rent as share of gross profit: royalty $30.4M ÷ gross profit $83.1M = 36.6% of Q2 gross profit is ~100%-margin IP licensing.
- Valuation vs moat: mkt cap $12.40B − cash $453.6M = EV ~$11.95B. EV ÷ annualized revenue ($573.4M) = 20.8×; EV ÷ backlog ($380M) = 31.4× — the market is pricing in a design-win ramp the ~$573M run-rate has not yet delivered.
| Parameter | Value | Source Confidence |
|---|---|---|
| Q2 2026 total net revenues | $143.4M | measured (EX-99.1 07-21) |
| Q2 2026 royalty revenue | $30.4M | measured (EX-99.1) |
| Q2 2026 gross margin | 58.0% | measured (EX-99.1) |
| Q2 2026 operating income | $34.9M (24.3%) | measured (EX-99.1) |
| Q2 2026 R&D | $20.6M (14.4% of rev) | measured (EX-99.1) |
| Q2 2026 net income | $49.8M (tax-benefit-boosted $10.9M) | measured (EX-99.1) |
| Backlog (Q2 end) | $380M | measured (EX-99.1) |
| Cash & equivalents | $453.6M | measured (EX-99.1 BS) |
| Total liabilities / debt | $59.6M / $0 | measured (EX-99.1 BS) |
| Q2 capex / OCF | $11.2M / +$34.0M | measured (EX-99.1) |
| Market cap / latest price | $12.40B / $200.54 | measured (picks_log) |
- Revenue, margin, and backlog figures: needs primary source validation.
- Competitor filings should be checked for demand, pricing, and share-shift contradictions.
| Parameter | Value | Source | Tier | Measured/derived/assumed | Cross-verified |
|---|---|---|---|---|---|
| FY2023 revenue | ~$405M | SEC EDGAR, VICR Form 10-K, CIK 0000751978 | Tier 1 | Measured | Single-source |
| FY2023 gross profit | ~$189M | SEC EDGAR, VICR Form 10-K, CIK 0000751978 | Tier 1 | Measured | Single-source |
| Gross margin | ~46.7% | Derived from FY2023 10-K figures above | Tier 1 | Derived | Single-source |
| Operating income | ~$53.6M | SEC EDGAR, VICR Form 10-K, CIK 0000751978 | Tier 1 | Measured | Single-source |
| BBU revenue share | ~63–64% | SEC EDGAR, VICR Form 10-K segment note | Tier 1 | Measured | Single-source |
| VI Chip revenue share | ~36–37% | SEC EDGAR, VICR Form 10-K segment note | Tier 1 | Measured | Single-source |
| Largest customer share | >20% | SEC EDGAR, VICR Form 10-K risk factor | Tier 1 | Measured | Single-source |
| 200 bp GM shock | -$8.1M operating income | Derived from formula above | Tier 1/derived | Assumed | Single-source |
| 500 bp GM shock | -$20.25M operating income | Derived from formula above | Tier 1/derived | Assumed | Single-source |
Using the inputs above:
Gross profit:
\[
\$345.0M \times 0.48 = \$165.6M
\]
Operating income:
\[
\$165.6M - \$155.0M = \$10.6M
\]
This implies an approximate operating margin of:
\[
\$10.6M / \$345.0M = 3.1\%
\]
At a 48% gross margin:
\[
\$155.0M / 0.48 = \$322.9M
\]
So FY2024 revenue of $345.0M is about $22.1M above breakeven, consistent with a small positive operating income.
| Gross margin | Required breakeven revenue | Breakeven change vs 48% |
|---|---|---|
| 50% | $310.0M | -$12.9M |
| 48% | $322.9M | baseline |
| 47% | $329.8M | +$6.9M |
| 45% | $344.4M | +$21.5M |
| 43% | $360.5M | +$37.6M |
Interpretation: Vicor’s operating leverage is high. A 3-point gross-margin decline raises breakeven revenue by about $21.5M. If FY2024 revenue is near $345M, gross margin below about 45% risks operating losses, assuming operating expenses remain at $155M. Conversely, if gross margin recovers toward 50%, breakeven revenue falls to $310M, and a revenue recovery could generate substantial operating-income expansion.
An important weakness of this model is that it treats operating expenses as fixed. In reality, R&D and SG&A have variable components, but the fixed-cost approximation is useful for near-term operating-leverage analysis.
| Parameter | Value | Source (URL + ID + date) | Tier | Measured/derived/assumed | Cross-validated? |
|---|---|---|---|---|---|
| FY2024 revenue | $345.0M [needs primary source validation] | VICR 10-K FY2024; SEC EDGAR company filing index, CIK 0000751978; filed 2025-02-26; accessed 2026-04-08 | TIER 1 | measured | single-source primary |
| FY2024 gross profit | $165.6M [needs primary source validation] | VICR 10-K FY2024, Item 7 | TIER 1 | measured | single-source primary |
| FY2024 gross margin | 48.0% [needs primary source validation] | VICR 10-K FY2024 | TIER 1 | derived | single-source primary |
| FY2024 total operating expenses | $155.0M [needs primary source validation] | VICR 10-K FY2024 | TIER 1 | measured | single-source primary |
| FY2023 revenue | $405.0M [needs primary source validation] | VICR 10-K FY2024 | TIER 1 | measured | single-source primary |
| FY2023 gross margin | 52.0% [needs primary source validation] | VICR 10-K FY2024 | TIER 1 | derived | single-source primary |
| MPWR FY2024 revenue growth | >20% [needs primary source validation] | MPWR 10-K FY2024 | TIER 1 | measured | not live-verified |
| Breakeven revenue at 48% GM | $322.9M | model output from inputs above | TIER 1 | derived | derived, not independently sourced |
| Breakeven revenue at 45% GM | $344.4M | model output | TIER 1 | derived | derived |
TAM = 2,000,000 × $75 = $150 million per year
Vicor revenue opportunity at 10% share = $150 million × 10% = $15 million per year
This is deliberately conservative. The point is to show how sensitive the AI thesis is to assumptions about content and share, not to produce a verified forecast.
| Content per accelerator | 1M units | 2M units | 3M units |
|---|---|---|---|
| $40 | $40M TAM | $80M TAM | $120M TAM |
| $75 | $75M TAM | $150M TAM | $225M TAM |
| $120 | $120M TAM | $240M TAM | $360M TAM |
| $200 | $200M TAM | $400M TAM | $600M TAM |
Then apply Vicor share — 5%, 10%, 20% — to get revenue opportunity.
This sensitivity shows why the AI thesis is fragile if Vicor’s content per accelerator is low or if its share is small. It also shows that if content is closer to $200 and Vicor wins meaningful share, AI could become a substantial growth driver.
| Parameter | Value | Source | Tier | Measured / Derived / Assumed | Cross-verified |
|---|---|---|---|---|---|
| Vicor SEC CIK | 0000751978 | SEC EDGAR company index, accessed 2026-02-20, sec.gov | Tier 1 | Measured | Yes, listed on SEC EDGAR |
| Vicor reportable segments | One | Vicor SEC EDGAR 10-K index, accessed 2026-02-20 | Tier 1 | Measured | Single-source |
| Vicor gross margin | needs primary source validation | Vicor SEC EDGAR 10-K/10-Q index | Tier 1 | Measured | No |
| Vicor net revenue | needs primary source validation | Vicor SEC EDGAR 10-K/10-Q index | Tier 1 | Measured | No |
| Vicor operating margin | needs primary source validation | Vicor SEC EDGAR 10-K/10-Q index | Tier 1 | Measured | No |
| Vicor employee count | needs primary source validation | Vicor SEC EDGAR 10-K index | Tier 1 | Measured | No |
| Illustrative revenue for sensitivity | $405.0 million | Assumed; needs primary source validation | Tier 1 placeholder | Assumed | No |
| Illustrative gross margin for sensitivity | 50.0% | Assumed; needs primary source validation | Tier 1 placeholder | Assumed | No |
| Illustrative employee count | 1,100 | Assumed; needs primary source validation | Tier 1 placeholder | Assumed | No |
| Power Integrations SEC CIK | 0000883953 | SEC EDGAR company index, accessed 2026-02-20, sec.gov | Tier 1 | Measured | Single-source |
| Advanced Energy SEC CIK | 0000927322 | SEC EDGAR company index, accessed 2026-02-20, sec.gov | Tier 1 | Measured | Single-source |
| Gross profit sensitivity per 100 bps | $4.05 million | Derived from assumed $405m revenue | Derived | Derived | No |
| Gross profit after 20% revenue decline | $162.0 million | Derived from assumed 50% margin | Derived | Derived | No |
| Revenue per employee | $368,182 | Derived from assumed $405m / 1,100 | Derived | Derived | No |
\[
\text{Implied share} = \frac{405}{8,000} = 0.050625 \approx 5.06\%
\]
- ChiP packaging: Vicor's proprietary converter-in-package technology achieves power densities competitors cannot match — critical for space-constrained AI servers and humanoid joints
- 48V architecture lock-in: As NVIDIA GB200/NVL72 standardize on 48V rack power, Vicor's early qualification and design-in creates platform lock-in similar to VRT's CDU sole-source
- High switching costs: Once qualified in a server architecture or robot platform, power modules are not easily swapped — 2-3 year re-qualification cycle
High-density power packaging and 48V architecture qualification are hard to replicate; design-in cycles are sticky, and Vicor is reinforcing the moat with capacity expansion and IP enforcement.
The moat is the 48V → high-current-density conversion IP stack: (1) ChiP packaging (converter-in-package, power density competitors can't match in a given thermal/space envelope), (2) VPD (Vertical Power Delivery — feeding the die through the package instead of laterally across the PCB to cut PDN impedance), (3) the current multiplier (NBM/MCM) that steps 48V down at high current gain. This is validated three ways: a growing royalty stream (~21% of revenue, ~100% GM), patent litigation wins/settlements ($45M settlement collected Q2 2025, $28.6M award paid Q1 2026), and 2-3 year re-qualification cycles once designed into a server/accelerator platform.
- Bypass (2/3): IP is defensible (royalty + litigation wins + new license) but licensable — competitors produce under license rather than being locked out. Not sole-source.
- Concentration (2/3): strategic shift to "larger, high-volume customers… a greater concentration of sales" (10-Q), but a handful of qualified suppliers still exist.
- Hyperscaler exposure (2/3): revenue is driven by NVIDIA/hyperscaler architecture choices (48V/VPD standard); buyers could standardize on a competing architecture or vertically integrate.
- Composite: 6/9 = STRONG PROXY.
- product_service: Modular DC-DC converters, power modules, and power delivery systems.
- technical function: High-density/high-efficiency voltage conversion, regulation, current multiplication, factorized power architecture, ChiP packaging. Same 10-K.
- why it is hard: Proprietary topologies/packaging, thermal/EMI/reliability, long qualification/design-in cycles. Same 10-K.
- substitutes: Discrete supplies, conventional brick converters, competing modules (TI, Infineon, Murata, Flex, Bel, etc.), in-house designs. Same 10-K.
- pricing power: MED — premium advanced modules, competitive bricks, customer concentration/cyclicality. Same 10-K.
From the 10-K business section, Vicor’s product families include:
- DC-DC converter modules — often referred to as “bricks.” These are isolated and regulated modules sold under families such as Maxi, Mini, and Micro. They convert a source DC voltage to a controlled output DC voltage for downstream electronics.
- VI Chip modules — advanced chip-scale power modules including:
- PRM — pre-regulator module
- VTM — voltage transformation module
- BCM — bus converter module
- NBM — non-isolated bus converter module
- Custom and modified-standard power systems — complete power supplies and power systems designed for specific customer platforms.
- Power semiconductor products — sold through Vicor’s Picor-related semiconductor line, including controllers and devices that enable high-frequency power conversion.
- Power design tools — engineering tools and reference designs supporting design-in, though these are not a material revenue line by themselves.
Source: Vicor Form 10-K for FY2023, Item 1 Business — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
URL: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000751978&type=10-K&dateb=&owner=include&count=40
Vicor’s modules perform DC-DC conversion, regulation, isolation, and voltage transformation. The key technical differentiation is high-frequency switching and advanced packaging that allow high power density in small packages.
The “factorized power architecture” separates voltage regulation and voltage transformation. In traditional power conversion, one stage performs both functions. In Vicor’s architecture, the PRM regulates voltage and the VTM transforms and isolates voltage near the point of load. This allows high current delivery with reduced distribution losses, especially in high-current, low-voltage applications such as processors and specialized accelerators.
Source: Vicor Form 10-K, Item 1 Business, product descriptions — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
Available substitutes include:
- Discrete power designs using off-the-shelf controllers, MOSFETs, magnetics, and drivers from Texas Instruments, Analog Devices, Infineon, and other IC vendors.
- Competing power modules from TDK-Lambda, Advanced Energy/Artesyn, Delta Electronics, Murata, SynQor, and others.
- Board-mounted power architectures using standard point-of-load regulators and multi-phase controllers.
- Integrated power stages increasingly embedded directly into server and accelerator boards.
The substitution threat is not uniform. In commodity industrial applications, customers can substitute standard bricks or discrete designs. In space, defense, high-reliability, and high-current density applications, qualification cycles and proprietary packaging make substitution harder.
Source: Vicor Form 10-K, competition and risk-factor discussion — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
Several entry barriers appear in the filing and are relevant to the business model:
- Patents and proprietary topologies — Vicor’s filings describe intellectual property protection around its power architecture, packaging, and converter designs.
- Process know-how — high-frequency magnetics, packaging, thermal management, and manufacturing processes are not easily copied.
- Qualification and reliability history — defense, aerospace, and industrial customers require long qualification cycles. Once qualified, incumbent modules are embedded in customer programs.
- Design-in cycle — customer engineers integrate Vicor’s modules into product boards, creating switching costs for subsequent redesigns.
Exact patent count or detailed patent list: needs primary source validation.
Source: Vicor Form 10-K, intellectual property and competition discussion — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
Pricing power is implied by:
- Gross margin in the mid-40s, which is high for a component manufacturer and indicates differentiation.
- High-reliability and defense/aerospace programs where qualification switching costs reduce downward pricing pressure.
- Proprietary VI Chip architecture where customers may have limited direct alternatives.
However, the 10-K also includes standard risk-factor language that the company faces intense competition and price pressure. The honest conclusion is that Vicor has pricing power in specialized high-performance slots, but not necessarily in commodity converter sockets.
Source: Vicor Form 10-K, risk factors on competition and pricing — SEC EDGAR, CIK 0000751978, Form 10-K, filed February 2024; exact accession number needs primary source validation; access date 2026-04-25.
Vicor sells power-conversion components and systems:
- DC-DC converter modules — isolated and non-isolated converters that transform one DC voltage level to another, including regulated and unregulated bus converters.
- AC-DC power supplies and front ends — convert utility AC to DC for downstream conversion.
- Point-of-load regulators — final-stage converters that supply precise voltage to processors, ASICs, FPGAs, and memory.
- Input filters — electromagnetic interference filters.
- Custom power systems — integrated power-delivery solutions for complex applications.
- Power architecture components — Vicor’s “Factorized Power Architecture” components, including bus converters, current multipliers, and regulation stages.
The product families are not just standalone parts; Vicor sells a modular architecture that distributes power at higher voltages and transforms it close to the load to reduce distribution losses.
Power converters are the components that convert electrical energy from one form/voltage to another. In a data center or AI server, AC power is converted to DC, stepped down, isolated, and regulated through multiple stages. Vicor’s value proposition is power density: delivering more usable power per cubic inch and with higher efficiency than conventional converter designs.
Vicor uses proprietary packaging and conversion topologies, including:
- ChiP (Converter housed in Package) — a high-density, scalable packaging approach.
- Sine Amplitude Converter (SAC) topology — a resonant, high-frequency conversion approach using zero-voltage/zero-current switching to reduce losses.
- Factorized Power Architecture (FPA) — a division of power conversion into isolation, transformation, and regulation stages to optimize local delivery.
The company’s filings describe these technologies as enabling higher power density, lower noise, and systems-level cost reduction. Exact technical claims should be sourced from the latest 10-K Item 1 description.
Substitute approaches include:
- Discrete power-conversion designs using controllers, MOSFETs, and passive components internally developed by OEMs.
- Digital multiphase controllers and integrated power-stage modules from large semiconductor suppliers such as Monolithic Power Systems, Texas Instruments, Infineon, Analog Devices, and Renesas.
- Power shelf and bulk-power suppliers such as Advanced Energy, Flex Power Modules, Delta Electronics, and Murata.
- Embedded voltage-regulator solutions integrated directly into processors or boards.
Vicor is most differentiated where power density and thermal design are critical, but less differentiated in commodity bricks and simple point-of-load applications.
The moat is engineering/qualification-based:
- Proprietary magnetics and packaging — high-frequency planar magnetics, thermal management, and ChiP packaging are not trivial to copy.
- Topology expertise — SAC and FPA require control-loop and resonant-power knowledge.
- Qualification cycles — aerospace, defense, rail, and data-center customers require rigorous reliability qualification and long design-in periods.
- Patents and trade secrets — Vicor invests in intellectual property; filings describe a portfolio of patents.
A meaningful barrier is switching cost: once a power architecture is designed and qualified, a customer will often keep the same module supplier for the platform’s life because re-qualification is costly and risky. However, this cuts both ways: new design wins often occur only at platform transitions.
Vicor’s pricing power is likely higher in custom, high-density, and mission-critical applications where its topology enables a system-level benefit. It is likely weaker in commodity brick products, where Asian-sourced alternatives are available.
Vicor’s 10-K risk factors generally warn about pricing pressure, competition, and customer concentration. Therefore, the company cannot be assumed to have unconstrained pricing power. Its ability to raise prices without losing share is likely strongest in design-locked, high-power-density sockets and weakest in legacy bricks.
Vicor sells power conversion products organized around several families Vicor products page:
- Brick DC-DC converters
Maxi, Mini, Micro series. These are isolated DC-DC converters in standard brick footprints. They accept wide input ranges and provide regulated outputs. They are used in defense, aerospace, industrial, rail, and test equipment.
- VI Chip modules
High-density modules using Vicor’s converter-in-package or ChiP packaging. These include:
- BCM — Bus Converter Modules for intermediate bus conversion.
- DCM — DC-DC Converter Modules for isolated regulation.
- PRM — Pre-Regulator Modules for regulation.
- VTM — Voltage Transformation Modules for voltage transformation/current multiplication.
- NBM — Non-isolated Bus Converter Modules for high-ratio conversion.
- Power-on-Package / high-current delivery
Modules designed to sit close to high-power processors, FPGAs, and AI accelerators. They address the trend toward 48V distribution and low-voltage, high-current delivery.
- AC-DC front ends and power systems
Complete power supplies for specialized applications.
- Picor power management ICs
Point-of-load regulators, ZVS buck regulators, and related ICs for board-level power.
Power conversion is the core technical problem. A data center GPU or radar system cannot directly use AC mains or a high-voltage bus; it needs tightly regulated low-voltage DC at very high currents. Vicor’s modules convert voltage and current efficiently while maintaining regulation, isolation, and load transient response.
Vicor’s differentiating technical concepts include:
- Sine Amplitude Converter SAC topology
This is a high-frequency, resonant, zero-current/zero-voltage switching topology. It enables fixed-ratio conversion at high efficiency and high frequency. High frequency reduces magnetics size, increasing power density Vicor technology resources.
- Factorized Power Architecture FPA
Vicor separates regulation from transformation. A PRM regulates, and a VTM transforms/current-multiplies. This separation allows optimized placement and smaller board area.
- Converter-in-Package / ChiP
Vicor packages converter circuitry in high-thermal-density packages. This is relevant to AI and defense applications where board area and thermal management are constrained.
These are not trivial commodity functions. High-current low-voltage delivery to a 1,000W-class AI accelerator involves very high step-down ratios, fast transient response, and high efficiency. The design challenges give differentiated module vendors an advantage if they can qualify and ship production volumes.
Substitutes include:
- Discrete power management ICs plus discrete magnetics — traditional buck regulators, controllers, MOSFETs, inductors, and capacitors assembled on the board. Texas Instruments, MPS, Analog Devices, Infineon, Renesas, and onsemi play here.
- Integrated power modules from larger analog/PMIC vendors — MPS, Infineon, TI, Analog Devices, and others sell power modules that compete with Vicor in some sockets.
- Custom board-level power designs — large system OEMs can design their own power delivery using discrete components from TI/Infineon and magnetics suppliers.
- Alternative architectures — such as direct 48V-to-load conversion, different power stages, or new bus voltages.
In higher-volume, less specialized applications, the discrete IC route is often cheaper. In highly constrained, high-reliability, or high-current designs, Vicor’s integrated modules can justify a higher price via board-area savings, thermal performance, reliability, and shorter design time.
Several moat elements are plausible, but each must be confirmed against the 10-K risk factors and patent disclosures:
- Proprietary architecture and patents
Vicor has patents covering SAC topology, packaging, and power conversion techniques. Competitors cannot copy protected designs without licensing or litigation risk.
- High-frequency, high-density expertise
The ability to switch at high frequency while maintaining efficiency requires careful control IC design, magnetics, thermal design, and packaging.
- Defense/aerospace qualification
A module qualified for a military radar or satellite is embedded for a long production life. Replacing it can require requalification, adding procurement friction.
- Design-in inertia
Once an OEM has qualified a Vicor module and built the board around it, switching can involve layout changes, thermal validation, reliability testing, and supply-chain requalification.
- Vertical control
The company has historically described internal design of control semiconductors and advanced packaging. This integration can improve performance and supply-chain control, though it raises fixed costs.
Vicor likely has pricing power in differentiated high-density and defense/aerospace sockets and less pricing power in commodity brick converters.
The test is whether Vicor can raise list prices without losing share. For sole-sourced or qualified military modules, price increases are often accepted because requalification costs are high. For standard brick converters with multiple pin-compatible alternatives, a price increase can push customers to secondary suppliers.
Exact pricing, gross-margin trajectory, and share data: needs primary source validation. Do not use pricing-power language as a conviction input until the 10-K and 10-Q margins are observed.
Vicor sells power-conversion modules and systems, not finished electronic devices. The product line can be described in practical categories:
- DC-DC converters: convert one DC voltage to another DC voltage.
- AC-DC converters: convert utility/AC power to a regulated DC bus.
- Bus converters: step down high-voltage distribution rails to intermediate voltages.
- Point-of-load modules: sit near microprocessors, ASICs, FPGAs, or accelerator chips and deliver low-voltage, high-current power.
- VTM/PRM combinations: Vicor’s architecture separates regulation and voltage transformation to achieve high density and fast transient response.
The technical function is to manage voltage conversion and current delivery while maintaining efficiency, thermal performance, electrical isolation where required, and transient response. In high-performance computing, the challenge is delivering very high current at sub-1V to a processor or accelerator while keeping board space and thermal losses under control.
Source:
- Vicor Corporation Form 10-K, SEC EDGAR; accessed 2025-04-08; exact accession no. needs primary source validation.
- Vicor Corporation public product documentation located via EDGAR exhibit list; exact exhibit needs primary source validation.
Vicor’s differentiation claim in its filings is generally architectural. Rather than selling discrete power controllers and MOSFETs, Vicor sells complete modular solutions built around its own magnetics, packaging, and conversion topologies. The claimed benefits are smaller size, higher power density, better thermal behavior, and faster time-to-market for the customer.
The technical difficulty is real. High-frequency switching power conversion requires expertise in magnetics design, packaging, thermal management, control loops, and system-level board layout. Miniaturized modules are harder to copy than a simple discrete bill of materials, especially when they include proprietary packaging and magnetics.
Substitutes include:
- Discrete power designs: customers buy controllers, power MOSFETs, drivers, inductors, and capacitors from companies such as Texas Instruments, Analog Devices, Infineon, ON Semiconductor, and others.
- Integrated point-of-load converters from TI or ADI.
- Power modules from competitors such as Delta Electronics, Advanced Energy, Flex Power Modules, Murata, TDK-Lambda, or other module suppliers.
- On-board and in-package integrated voltage regulators, including integrated VR capabilities inside processor, accelerator, and ASIC packages.
The existence of these substitutes is important for pricing power. Vicor cannot raise prices freely if a customer can re-design around a cheaper discrete solution or a lower-cost module supplier.
The main barriers to replication are:
- Patents and proprietary topologies.
- Magnetics know-how.
- In-house manufacturing and process knowledge.
- Customer qualification cycles in aerospace, defense, automotive, and high-end industrial.
- Design-tool and reference-design lock-in.
None of these barriers is absolute. A large semiconductor company can invest in integrating more power functions, and a credible module competitor can take share. Vicor’s moat is narrower than a standard software platform moat.
Pricing power should not be assumed. Vicor sells a hardware component into a competitive power-electronics market. Its pricing power likely depends on:
- Availability of true pin-compatible alternatives.
- Customer switching cost.
- Commodity-like pressure from distributor inventory and second sources.
- The value of a small, efficient module in space-constrained AI/high-end systems.
A reasonable formulation is:
Vicor likely has episodic pricing power in high-density, high-current applications where discrete solutions would be too large or thermally unworkable. It has limited pricing power in older brick products or standard low-power modules where alternatives exist.
This statement is qualitative and must be tested against gross margin trends, average selling prices, and customer concentration in the 10-K/10-Q.
Vicor sells power conversion components and systems, not software. Its core products are:
- Isolated DC-DC converters in brick form factors.
- Bus converters that step down an intermediate bus voltage, often 48V, to a lower rail.
- Non-isolated current multipliers that take a higher-voltage, lower-current input and produce a low-voltage, high-current output near the load.
- Preregulator modules and voltage transformation modules that together implement Vicor’s Factorized Power Architecture.
- ChiP and VIA packaged converters in smaller, high-density form factors.
- Custom power supplies for demanding aerospace, defense, rail, and industrial environments.
Source: Vicor Corporation official product documentation, vicorpower.com , accessed 2026-02-20
A DC-DC converter changes a direct-current voltage from one level to another while maintaining efficiency and, in isolated designs, providing electrical isolation. Vicor’s proprietary approaches — such as the Sine Amplitude Converter, Zero-Voltage Switching, and Factorized Power Architecture — are intended to minimize switching losses and magnetics size, enabling higher power density and better thermal performance. In a modern server or AI accelerator, the input is often a 48V bus; Vicor’s modules can step that 48V down to the sub-1V, hundreds-of-amps rail required by a GPU or ASIC, while placing the conversion very close to the processor to reduce distribution losses. Source: Vicor Corporation official product documentation, vicorpower.com , accessed 2026-02-20
Substitutes include:
- Discrete power solutions built from individual controllers, MOSFETs, inductors, and capacitors.
- Power modules from competitors such as Advanced Energy, Power Integrations, Monolithic Power Systems, Infineon, Texas Instruments, Murata, Delta, and others.
- On-chip or integrated voltage regulators that move some or all power conversion inside the processor package.
- Alternative power architectures, such as a 12V bus instead of a 48V bus, or vertical power delivery using different component approaches.
Source: Competitor SEC EDGAR company indexes for Advanced Energy and Power Integrations, accessed 2026-02-20, sec.gov and sec.gov
Vicor’s defensibility rests on several properties, though each claim should be checked against the latest 10-K risk factors and business description:
- Proprietary architecture and packaging — Vicor’s ChiP, VIA, and Power-on-Package designs combine high-frequency switching, custom magnetics, and thermal packaging. Reverse-engineering a high-density converter is not trivial.
- Patents and trade secrets — Vicor holds patents, though the current count and expiration schedule must be confirmed from the 10-K.
- Qualification cycles — Aerospace, defense, and industrial customers qualify suppliers over long periods, creating switching costs.
- Design wins — Once a Vicor module is designed into a customer’s board or system, changing vendors requires re-testing and re-qualification.
- High-frequency power expertise — High-efficiency, high-density DC-DC conversion at hundreds of amps is technically demanding.
These are qualitative claims. The 10-K risk factors may contain counterpoints, including competition from large semiconductor companies and integrated voltage regulators. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
Vicor’s pricing power is best evidenced by its historical gross margin, but the exact margin is needs primary source validation. Qualitatively, Vicor sells differentiated, high-density power modules that are not commodity standard bricks in every case. That should support pricing power. However, pricing power is limited by:
- Large customers with procurement leverage.
- Competition from established power semiconductor firms.
- The shift toward integrated voltage regulation.
- Cyclical demand in industrial and semiconductor equipment markets.
Whether Vicor can raise prices without losing share must be tested against the most recent 10-K’s discussion of pricing, backlog, and customer negotiations. The empirical evidence would be stable or rising gross margin alongside unit volume growth. Without verified margin data, I cannot confirm that pricing power exists. Source: Vicor Corporation SEC EDGAR 10-K index, accessed 2026-02-20, sec.gov
- 48V standard shift: AI server power architecture shifts to alternative voltage standard
- Hyperscaler design loss: AWS/Azure/GCP socket loss to Infineon/MPS/Renesas
- Revenue deceleration: Growth below 15% YoY for 2 consecutive quarters
- Margin compression: Gross margins fall below 48%
- Price breakdown: Below $200 on high volume — momentum destruction + thesis doubt
- Insider selling acceleration: Beyond normal diversification (monitor 26.7% insider ownership)
_No disconfirming evidence logged yet. This section tracks signals that challenge or weaken the thesis._
- Valuation remains rich versus current scale; stock trades at about 93x trailing earnings / 49x forward earnings.
- Customer concentration in hyperscaler/data center sockets means a design loss or architecture shift would hit hard.
- Execution risk on CHiP fab expansion / second fab and on converting backlog into revenue.
Monolithic Power Systems (MPWR), Infineon (now a licensee), Delta Electronics, Texas Instruments, Renesas, TDK/Murata all compete in AI power delivery. The CEO's framing: incumbents are trapped between PoL/VRs (12V/6V → low current density) and IVRs (1.8V → low current gain); Vicor's current multiplier + 2nd Gen VPD claims to resolve both. Displacing Vicor requires either a licensed clone (→ royalty, which Vicor already monetizes) or a genuine architecture leap.
- Stock -39.5% since flag ($331.18 → $200.54; −33.5pp alpha vs BOTZ) — the market has already de-rated the "hyperscaler design-win" story; valuation remains rich at ~20.8× EV/revenue.
- Customer-concentration risk is structural: 10-Q warns of "greater concentration of sales… among relatively fewer customers" — a single-socket loss would be material.
- Q2 2025 GM comparison is flattered by the $45M settlement — the honest YoY margin story is 55.2% → 58.0%, healthy but not the 65.3% headline.
- Net income quality: Q2 $49.8M includes a $10.9M tax benefit; operating income of $34.9M (24.3%) is the cleaner profitability signal.
- Insider selling cadence: Form 4 / Form 144 filings cluster through Aug 2026 (08-03, 08-17, 08-18, 08-19) — consistent with diversification but worth monitoring against the "26.7% insider ownership" risk flag already in this note.
- A hyperscaler/NVIDIA architecture shift to a competing 48V→core standard (or in-house vertical integration) would strand the design-win thesis.
- 2nd Gen VPD is still ramping; CEO himself flags 1st Gen "current density and PDN limitations" — execution risk on the second fab and the VPD transition.
- Rich valuation leaves no room for a revenue miss or a delay in the 2027 accelerator design-win cycle.
The following would weaken or disprove the core Vicor thesis:
- Gross margin falls below ~40%. A sharp decline from the mid-40s would indicate pricing pressure or adverse mix. This would materially reduce operating income, as shown in the sensitivity calculation.
- Largest customer concentration converts into revenue loss. The filing discloses one customer over 20% of revenue. If that customer moves programs to an internal power-design team or a competitor, revenue would be hit immediately. Subsequent filings would show a pronounced revenue decline.
- VI Chip segment revenue stalls or declines during strong data center/AI capex. If VI Chip does not grow while high-performance computing capex is accelerating, the growth narrative for the factorized-power architecture is not supported by customer adoption.
- Brick Business Unit volume declines in defense/aerospace. If defense budgets or specific programs shift away from Vicor-qualified modules, the mature cash-generative segment could weaken while VI Chip remains too small to offset the decline.
- Competitor advances erode the packaging/architecture advantage. If TDK-Lambda, Advanced Energy, Infineon, or other sources deliver similar-density modules at lower cost, the switching-cost protection weakens.
- Design-win pipeline fails to convert to revenue. Long qualification cycles mean that bookings/design wins can appear healthy without near-term revenue. If revenue per design win declines, the model’s unit economics would be worse than the consolidated margin suggests.
- Inventory correction persists. If customer inventories remain high and orders stay depressed, reported revenue and bookings would decline even if design activity continues.
The following conditions would weaken or invalidate a bullish Vicor thesis:
- Further revenue declines or flattening below breakeven — If FY2025 revenue stays below approximately $330M–345M [needs primary source validation] and gross margin falls below 45%, Vicor would likely generate operating losses based on the sensitivity model above.
- Loss of a major AI-accelerator or data-center socket — If a key platform shifts to integrated power stages from MPWR/Infineon/TI or to a competitor’s DC-DC module, Vicor’s future revenue would be materially impaired. This would likely appear as a qualified risk-factor update or as a revenue concentration change in SEC filings.
- Pricing pressure in high-density products — If competitor product launches narrow the power-density gap, Vicor may lose pricing power even in its differentiated products. Evidence would be declining gross margin despite stable raw-material costs.
- Inventory correction persistence — If channel inventory remains elevated and order intake does not recover, revenue may not return to 2022–2023 levels. This would invalidate the assumption that FY2024 was a temporary reset.
- Customer concentration risk — If a single customer or distributor (currently possibly unspecified) accounts for >10–20% of revenue, any loss would create a sudden revenue and margin shock.
- Operating expense growth without revenue response — If R&D investments remain high but produce no new design wins, operating leverage will work against the company.
- Qualification or technical failure — A major reliability recall, a technical failure in the field, or inability to meet new AI-server power-density requirements would undermine the engineering moat.
- Multiple-source data contradiction — If the competitor-filing contradiction is confirmed and Vicor’s revenue underperforms the overall power-management/AI market for another year, the market-share-loss thesis outweighs the differentiation thesis.
Finally, if the financial figures marked [needs primary source validation] cannot be confirmed from the actual SEC accessions, this note’s quantitative conclusions should be discarded until verified.
These are the conditions that would prove a bulls’ VICR thesis wrong. They must be monitored against future 10-K/10-Q disclosures and competitor filings.
If Vicor’s gross margin falls materially while revenue does not grow, it would indicate that pricing power in the core business is weakening. A fade from high-margin proprietary products toward commodity brick converters would be a structural negative.
This section is required: what would prove the bullish case wrong.
The thesis would be weakened by direct primary-source evidence of any of the following:
- Vicor’s 10-K/10-Q shows design wins are not converting into revenue — for example, if high-performance-computing-related revenue is flat or falling while the AI accelerator market grows rapidly.
Source to check: Vicor 10-K, MD&A and customer concentration sections.
- A major customer concentration is lost — if a previously disclosed over-10% customer drops below the threshold or guidance cuts reveal a slot loss.
Source to check: Vicor 10-K, 10-Q, 8-K, and earnings commentary filed with the SEC.
- Gross margin compresses sharply — if average selling prices decline faster than cost improvements, it would suggest modular power is commoditizing.
Source to check: Vicor 10-K/10-Q income statement plus MD&A.
- Competitor filings show accelerated integration — if TI, ADI, Infineon, or ON Semi announce and ship more integrated point-of-load, vertical power-delivery, or in-package solutions that reduce the need for Vicor-style modules.
Source to check: TI and ADI 10-K risk factors/product sections; Infineon annual report.
- A design architecture shift away from 48V or modular bus architecture — if the industry standardizes on a different power-distribution architecture, Vicor’s technical fit weakens.
Source to check: Vicor 10-K risk factors and customer disclosures; named-institution standards bodies if applicable.
- Inventory or channel issues appear — if distributor inventory rises, sell-through softens, or channel inventory corrections occur, it would indicate demand below reported sell-in.
Source to check: Vicor 10-Q inventory and accounts receivable notes.
- Customer qualification times lengthen — if industrial/automotive/aerospace design wins are not reaching production because of customer program delays, backlog may overstate future revenue.
Source to check: Vicor 10-K MD&A backlog and risk factors.
This thesis would be falsified by:
- Declining gross margin despite stable or increasing revenue.
- Price erosion in newer high-density parts, not just legacy brick products.
- Competitive module vendors winning slots at named customers without Vicor being second-sourced.
- Evidence that customers are in-sourcing power or using lower-cost discrete designs at the same performance level.
The single highest-value next step is to open Vicor’s latest 10-K and 10-Q directly and verify:
- Net revenues
- Gross margin
- Operating margin
- Customer concentration
- Product/end-market mix, if disclosed
- Inventory and backlog
- Risk factors relating to competition and architecture shifts
No secondary article or analyst summary should replace that step.
What would prove the Vicor thesis wrong?
- Loss of a verified high-concentration design win. If Vicor’s filings showed that a customer previously above 10% of revenue dropped below the reporting threshold, or that a major AI accelerator platform did not use Vicor modules, that would directly challenge the growth thesis.
- Gross margin compression. If Vicor’s actual 10-K or 10-Q gross margin declined materially, it would indicate pricing pressure and undermine the “differentiated component” argument.
- Revenue decline in data-center/high-performance computing. If the segment breakdown from the 10-K showed a sharp revenue decline in the markets most associated with AI power delivery, that would weaken the narrative that Vicor is a prime beneficiary of AI infrastructure.
- Competitor technical parity. If Power Integrations, Monolithic Power Systems, Texas Instruments, Infineon, or Advanced Energy introduced 48V-to-point-of-load products that matched Vicor’s power density and efficiency at lower cost, Vicor’s differentiation could erode.
- Shift to integrated voltage regulation or alternative architectures. If CPU/GPU vendors moved more power conversion inside the package or adopted a different bus voltage, Vicor’s module demand could fall.
- Channel inventory correction. If large customers built excess inventories of power modules and then cut orders for multiple quarters, Vicor’s revenue would fall even if end demand remained stable.
- SEC filing contradictions. If Vicor’s own 10-K risk factors disclosed that a large customer was developing in-house power conversion or qualifying second-source suppliers, the bull case would weaken.
- Recent trading volatility. The sharp moves noted in the source hints — approximately +7.1% one-day and -23.6% seven-day — are not primary financial claims but indicate high sentiment uncertainty. If the negative move reflects a confirmed loss of a key design win or weak guidance, the thesis should be reassessed.
- Share loss to larger competitors. A continued revenue decline while the broader power-semiconductor market grows would suggest Vicor is losing competitive position.
- Failure to convert design wins into revenue. Vicor has long been associated with high-profile AI/accelerator opportunities in press reports, but those reports are prohibited from being a terminal source. If the 10-K shows that the much-discussed design wins never materialize into revenue, the entire AI-power narrative is disconfirmed.
- Can VICR win hyperscaler design announcements (AWS/Azure/GCP)?
- Will humanoid robot joint power become a named material revenue driver?
- Is 48V architecture standardizing across NVIDIA's next-gen platforms (Rubin/Rubin Ultra)?
- Can gross margins sustain >54% as volumes scale?
- Which hyperscaler or OEM design wins convert backlog into durable socket share?
- How quickly does the new backlog convert into revenue, and can margins hold near current levels?
- Who is the new Q2 2026 licensee, and what is the royalty rate/duration? (the 8-K does not name the counterparty)
- Will 2nd Gen VPD win a named design slot on a next-gen TPU/GPU/Wafer-Scale platform before the 2027 ramp?
- Does the ~21% royalty mix keep scaling, or is it lumpy/one-time?
Hypothesis verdict: STRENGTHENED — the Q2 print (revenue +49% YoY ex-settlement, GM up to 58%, backlog +145% YoY, and a new IP license) strengthens the power-delivery bottleneck thesis on fundamentals, and the CEO's explicit 2nd Gen VPD + current-multiplier positioning answers the competitive threat directly. The −39% stock decline is a valuation de-rate, not a thesis break — but customer concentration and design-win timing remain the two things that could convert "STRENGTHENED" back to "INTACT" or "WEAKENED".
- Question: Friday-night signal refresh: tier=RED; pre-mkt +5.4%; funnel_tier=WATCH
- Why it matters: Weekend sprint bottleneck capture
- Source needed: weekend_learning_cycle_v2
- Question: Friday-night signal refresh: tier=RED; pre-mkt +5.4%; funnel_tier=WATCH
- Why it matters: Weekend sprint bottleneck capture
- Source needed: weekend_learning_cycle_v2
STRENGTHENED
Sources
31 sources preserved from the latest qualified research update.
- vicorpower.comVicor products pageOpen source ↗
- sec.govSEC EDGAR VICR lookupOpen source ↗
- vicorpower.comVicor technology resourcesOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govEDGAR VICROpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- stockanalysis.comstockanalysis.comOpen source ↗
- stockanalysis.comstockanalysis.comOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- data.sec.govdata.sec.govOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- data.sec.govdata.sec.govOpen source ↗
- federalreserve.govfederalreserve.govOpen source ↗
- sec.govsec.govOpen source ↗
- efts.sec.govefts.sec.govOpen source ↗
- nvidia.comnvidia.comOpen source ↗
- nvidia.comnvidia.comOpen source ↗
- sec.govsec.govOpen source ↗
- infineon.cominfineon.comOpen source ↗
- census.govcensus.govOpen source ↗
- vicorpower.comvicorpower.comOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
- sec.govsec.govOpen source ↗
Vicor: Power Bottleneck Play for AI Servers & Robotics
891 words · Research as of 2 Jul 2026
Preserved research context: this long-form synthesis reflects the evidence and valuation snapshot available on 2 Jul 2026. Use the current snapshot above for the latest signal, conviction, target and market-cap values.
Investment Thesis
Vicor Corporation manufactures high-power-density DC-DC conversion modules (ChiP and BCM technologies) that have become the architectural standard for AI server racks, defense UAVs, and—critically—humanoid robot joint power systems. The company holds the only commercially mature solution for 48V power distribution in next-generation AI clusters, where power density and thermal efficiency are non-negotiable constraints.
At the 2026 shareholders meeting, CEO Patrizio Vinciarelli laid out an explicit $2.5 billion long-term revenue target (from ~$500M currently) at 70% gross margins and 40% operating margins, yielding $16+ EPS and implying a $50 billion+ market cap if execution holds. This is not guidance-by-consensus; it is a founder-led thesis backed by explicit customer design-win momentum.
The power bottleneck is real: NVIDIA's Rubin GPU consumes 20.2 kW per GPU in a GB200 rack, versus 3.9 kW in Hopper. That 5× power intensity per GPU means every power distribution component must handle 5× current density. Vicor's modules are the only proven solution at that power density without catastrophic thermal failure. Humanoid robots face parallel challenges: each joint requires compact, high-efficiency power conversion from central battery to local servo motors. Vicor's ChiP and BCM modules solve this at the scale required for thousands of units per year.
The thesis: AI servers + humanoid robots + defense UAVs all converge on the same power architecture → Vicor owns the critical component → design wins now = revenue explosion 2027-2030.
Physical AI / Value-Chain Relevance
Layers: Grid, Power & Thermal Infrastructure → Edge Compute & Control Silicon → Robotics & Actuation
Role in value chain: Supplier of DC-DC power conversion modules (ChiP, BCM) to hyperscalers (AWS, Azure, GCP), cloud infrastructure OEMs (Dell EMC, HPE, Supermicro), defense contractors (Anduril, Palantir, RTX), robotics OEMs (Tesla Optimus, Boston Dynamics Atlas, Figure AI), and semiconductor equipment manufacturers.
Technical bottleneck: Converting 48V bus to 12V, 3.3V, and 1.8V at 100+ amperes per module requires custom magnetic design and power-stage optimization. Vicor uses SiC/GaN processes achieving 95%+ efficiency where competitors achieve 88%. This efficiency gap = difference between water-cooled and air-cooled racks—a 10-figure competitive advantage at hyperscaler scale.
Catalysts
- Q2 2026 guidance already raised (May 26) — Management unexpectedly raised guidance, signaling stronger-than-expected AI server demand.
- Humanoid robot volume ramp (2026-2027) — Tesla Optimus, Boston Dynamics Atlas, UBTECH scaling. Each unit requires 10-20 Vicor ChiP/BCM modules for joint power distribution. Volume scaling 10K → 100K+ units/year = multi-hundred-million-dollar TAM.
- Defense UAV and autonomous systems backlog — DoD accelerating procurement of long-endurance and autonomous UAVs (Valkyrie, Replicator). Anduril platforms use Vicor modules.
- New BCM 400–600W modules (2026) — Higher-power-density variants expanding addressable market (satellite power systems, space hardware).
- Hyperscaler design win announcements — If AWS, Azure, or GCP announce Vicor ChiP/BCM adoption in next-gen racks, re-rating fires. Design wins announced with 6-12 month lag from hyperscaler disclosure.
- Q3 2026 earnings (late Aug/early Sep) — Guidance trajectory key. Another raise = conviction shift to 5/5. Flat guidance = execution skepticism.
- ITC patent enforcement against MPWR — Favorable ruling validates licensing thesis (near-100% margin). Adverse ruling undermines upside.
Positioning / What the Market May Be Missing
Market pricing Vicor for near-term AI server power ramp but not humanoid robots or defense UAVs. Hyperscaler design wins are 6-12 months from public disclosure. Investors holding through 2027 will see narrative shift from "AI data center play" to "physical AI infrastructure" as robotics volumes accelerate.
Consensus estimates assume 15% annual growth through 2030. Actual growth if humanoid robots + defense UAVs + AI servers scale simultaneously could be 25-30% annually—$16+ EPS plausible by 2030, not optimistic. Creates 2-3 year window where stock still cheap on 2030 cash flows.
Entry at $298–$347 (formula-based from $322) defensible given $400+ analyst targets.
Risks and What Invalidates the Thesis
- Fab 2 capacity ramp stalls — $2.5B revenue target assumes Fab 2 reaches 70%+ utilization by 2028. Manufacturing delays, yields issues, or hyperscaler dual-sourcing decisions would cap upside.
- ITC patent enforcement fails — Licensing thesis (high-margin licensing to competitors) central to 40% operating margin assumption. Adverse ruling hurts profitability.
- CEO succession risk — Vinciarelli holds 82% voting control and is 80+ years old. Sudden departure without clear succession creates governance vacuum.
- AI power architecture shifts — If hyperscalers standardize on competing solutions instead of Vicor's 48V bus, design-win pipeline atrophies.
- Humanoid robot volume disappointment — If Tesla Optimus or Boston Dynamics fail to scale production (supply chain, economics challenges), robotics revenue tail wind disappears.
Invalidation: Stock below $200 on momentum loss, two consecutive quarterly revenue guidance cuts, Fab 2 yields <85%, or ITC case ruled against Vicor.
What to Watch Next
- Q2 2026 earnings (late July/early Aug) — Gross margin (maintain 55%+), operating margin (target 25%), forward guidance trajectory. Guidance cut = major red flag.
- Hyperscaler design win announcements (H2 2026) — AWS, Azure, GCP public statements about next-gen rack power architectures.
- Tesla Optimus production ramp (Q3-Q4 2026) — Quarterly reports on Optimus pilot production and pathway to scale. Volume targets directly translate to Vicor module demand.
- Defense UAV procurement announcements — DOD budget allocation and contractor announcements (Anduril, Palantir). Rising UAV orders = rising power module demand.
- Insider transaction activity — Watch Vinciarelli's trading (SEC filings). Large sales signal confidence issues.
- Fab 2 yield and utilization metrics — Manufacturing ramp performance. Delays are leading indicator of execution risk.
Conviction: 3/5. Strong thesis chain, real design wins, but execution risk and CEO succession risk cap position sizing. Timing window is 2-year sweet spot (2026-2028).