UmbriaExplore our travel guides. Share your stories, tips and questions on Reddit.
Fast movers • Alerts • Momentum
10% OFF with SAVE10
Charts • Watchlists • Multi-broker tools
Built for active traders
Biotech catalyst, news and analysis PDUFA tracker
Biotech catalyst, news and analysis PDUFA tracker
A semiconductor foundry, acquired security and sensing businesses, and a new computing roadmap expand the opportunity — and the execution, spending and dilution tests.
Get every Merlintrader report in real time on Telegram: join @merlintraderpub_com.
The October 12 Elisity announcement expects a fourth-quarter 2026 closing, subject to regulatory approval and other conditions.[6] Superion 256 customer deliveries are expected in 2027; FIU installation is expected in late 2027 and the NVIDIA research-center installation is scheduled for 2027.[8][12][13] These are announced windows, not completed milestones. The next financial statement will also show a new reporting perimeter after SkyWater closed on July 31.[2]
Elisity is an announced all-stock acquisition: $1.15 billion at closing, up to $150 million earnout, and a separate $100 million employee RSU retention pool.[5] SkyWater and Oxford Ionics are already closed acquisitions, but their benefits still depend on operating execution.[2][1] Contract ceilings and prototype results are not recognized revenue or a completed utility-scale computer.
Editorial scenario: manufacturing access improves the design cycle, acquired businesses broaden customer relationships, and computing orders become dependable deployments. SkyWater is closed and Superion prototypes have trapped their first ions; 2027 deliveries remain the customer test.[7][8] Substantial June financial resources provide development capacity, but returns depend on productive use of that capital.[2]
Editorial scenario: integration and technical scaling take longer while the cost base and ownership denominator expand. Q2 operating loss was $337.243 million and H1 operating cash use was $254.781 million.[2] Stock-financed acquisitions preserve cash but issue claims on future value. A larger consolidated business need not automatically create stronger per-share economics.
Q2 revenue was $80.050 million; the $1.867742 billion net loss attributable to IonQ included a $1.649115 billion warrant fair-value loss.[2] That mark-to-market expense is not same-sized quarterly cash consumption. June cash, equivalents and investments totaled a calculated $2.959334 billion before SkyWater; the August release gave a rounded $2.0 billion pro-forma balance after acquisition-related cash uses.[2][3]
IonQ now combines computing, networking, sensing, security, space-related services and semiconductor manufacturing.[1][7] Its operating question is whether that breadth creates repeatable customer economics. September guidance of $450–460 million includes SkyWater, unlike August guidance of $280–290 million; the difference is not an organic-growth figure.[4][3] DARPA Stage C, a one-kHz interconnect and error-correction experiments add specific evidence, while independent validation and commercial scaling remain ongoing tests.[11][17][18]
The $2.0 billion raise, earmarked for working capital, general corporate purposes and possible acquisitions or partnerships, gives IonQ a much larger cash cushion. The trade-off is a bigger share count: new shares and pre-funded warrants now, plus 100 million Series C warrants at $69.46 whose full exercise, worth roughly $6.95 billion, would happen only at higher prices, on top of older warrants and the stock earmarked for Elisity. The questions the market is weighing are whether sales growth through 2027 justifies the enlarged capital base and how management handles a business spread across several fronts.
IonQ closed on October 16, 2026 the offering underwritten by J.P. Morgan Securities as sole underwriter under an agreement of October 14: 20,500,000 common shares, 29,500,000 pre-funded warrants and 100,000,000 Series C warrants. Each share or pre-funded warrant was sold with two Series C warrants at a combined public price of $40.00, for gross proceeds of about $2.0 billion; the prospectus supplement estimates net proceeds of about $1,988.5 million, excluding any warrant exercise, for working capital, general corporate purposes and possible acquisitions and partnerships. The Series C warrants have an exercise price of $69.46 and, like the pre-funded warrants, are exercisable immediately for seven years. The prospectus puts common stock outstanding after the offering at 429,407,678 shares, or 558,907,678 if all the warrants are exercised. The warrants add potential dilution on top of the new shares. Form 8-K, prospectus supplement
Morgan Stanley kept its Equal-Weight rating on IonQ and cut its price target to $43 from $49, Dow Jones Newswires reported on October 15, 2026. A third-party analyst opinion, not a Merlintrader view and not a recommendation.
Source: Dow Jones Newswires, October 15, 2026.
IonQ and the University of Cambridge announced the opening of the IonQ Quantum Innovation Centre at the Ray Dolby Centre, home of the Cavendish Laboratory, turning the partnership first announced on March 11, 2026 into an operating research centre. According to the company, nine inaugural projects selected by a Joint Steering Committee pair Cambridge researchers with IonQ specialists across quantum computing, networking, sensing and security, from trapped-ion light-collection hardware to algorithms for chemistry, materials, finance, engineering and medicine. IonQ says it will deploy a Superion 256-qubit system and an on-campus entanglement-distribution network with two Quantum Networking Data Centres and three Quantum Networking Nodes, strengthening the existing Cambridge-to-Bristol quantum network, and Innovate UK will provide National Quantum Computing Centre access and computing time over three years. The release describes the centre as deepening IonQ’s presence in the United Kingdom but gives no deployment date and no investment or funding amount; the Superion installation is an announced plan, not a completed delivery. IonQ release
SkyWater and D-Wave announced a multiyear agreement for specialized process development and manufacturing at the Minnesota fab, with terms protecting D-Wave intellectual property.[22] The release does not disclose a contract value. This is external foundry work, not an IonQ computing benchmark.
IonQ announced a transaction valued at up to $1.3 billion of stock merger consideration, including an earnout.[6] The 8-K separately records a $100 million employee RSU pool and a VWAP collar of $33.07–$49.61 for the share calculation.[5]
IonQ reported more than 1,000 entanglement events per second between a trapped ion and a quantum memory.[9] The preprint reports 1.03(1) kHz and 87.9(7)% fidelity: a measured component result, not a completed quantum data center.[17]
IonQ entered QBI Stage C with potential agreement value up to $300 million; funding beyond the initial obligation depends on later appropriations and funding actions.[10] DARPA says the stage tests systems, not that utility scale is already proven.[11]
A Superion 256 is scheduled for installation in 2027, linked to NVIDIA infrastructure through NVQLink and CUDA-Q.[13] This is a deployment and joint-research plan, not an announced completed installation.
Superion 256 was introduced with 2027 customer deliveries expected.[8] Updated 2026 revenue guidance became $450–460 million including SkyWater from July 31 and estimated intercompany eliminations.[4]
The full deep dive has the answer’s building blocks: cash, dilution, catalysts and risks, every figure sourced.
Free. No signup. You decide, we don’t recommend.
IonQ now combines a quantum-computing development program, businesses with existing security, sensing and space-related customers, and a semiconductor foundry. Its June financial statements describe computing, networking, sensing, security and satellite-based capabilities; SkyWater became a subsidiary after that reporting period.[2][7] This is not a single commercial experiment. The central analytical question is whether the pieces can reinforce one another without making the cost base, management workload and ownership denominator grow faster than durable economic output.
In a constructive scenario, the foundry shortens the development cycle, acquired customer relationships open additional distribution channels, and computing deployments generate repeat orders rather than isolated research purchases. IonQ’s September Superion announcement reported initial chips fabricated at SkyWater and first ions trapped in prototype systems, while placing customer deliveries in 2027.[8] The favorable interpretation is that manufacturing access has begun to affect engineering execution. It would become stronger if those prototypes lead to accepted systems with dependable performance, serviceability and economics.
A middle outcome would combine rising consolidated revenue with a long period of expensive development. The September 8 outlook calls for $450 million to $460 million of 2026 revenue including SkyWater from July 31, whereas the August outlook excluded SkyWater.[4][3] The business could meet the higher consolidated range while the original computing operation develops at a less dramatic pace. That would not make the reported revenue fictional; it would make the composition of growth more important than the headline percentage.
A stock-financed acquisition can preserve cash yet still disappoint existing holders. The announced Elisity transaction would exchange IonQ shares for another business and add a separate employee retention pool.[5] If the acquired activity creates less lasting value than the ownership issued, consolidated revenue can expand while the economic claim of each existing share weakens. The scenario framework therefore keeps company scale, commercial validation and per-share economics separate. No scenario implies a price target or a recommendation.
The first test is delivery. Superion 256 was available to order in September 2026, with initial customer deliveries expected in 2027; FIU’s installation was expected in late 2027 after completion of its campus data center.[8][12] A dated order and an installation target are evidence of a commercial pathway, not a completed operating system. Confirmed installation, acceptance, uptime and usable workload results would move the evidence further along that pathway. Slippage accompanied by little technical detail would move it in the opposite direction.
The second test is independent validation. DARPA selected IonQ for Stage C of the Quantum Benchmarking Initiative, alongside other approaches, and describes that stage as testing whether proposed systems can be built and operated as intended.[11] This creates a demanding external evaluation channel. It does not pre-certify a utility-scale computer. The constructive interpretation strengthens if successive systems meet the milestones under review; it weakens if scaling assumptions fail when hardware is tested beyond the conditions of a small demonstration.
The third test is the quality of growth. IonQ defines organic revenue as revenue from its quantum-computing products and acquisitions completed before December 31, 2024.[3] Under that definition, revenue from later acquired businesses is outside the organic measure. A meaningful operating improvement would show organic growth together with clearer delivery economics, cash collection and repeat demand. Consolidated growth alone cannot show those points when the reporting perimeter has expanded substantially.
IonQ, Inc. is a Delaware corporation headquartered at 4505 Campus Drive, College Park, Maryland. Its common stock, with a par value of $0.0001 per share, trades on the New York Stock Exchange under IONQ.[5] The listed common share represents ownership of the consolidated company, not just its trapped-ion computing operation. That distinction became more important as IonQ purchased businesses serving security, sensing, space and semiconductor customers.
The annual report describes a quantum platform offering computing, networking, sensing and security, supported by custom hardware, firmware and an operating system.[1] It also describes satellite imagery and satellite solutions intended to contribute to quantum-secure communications.[1] Some of these activities sell existing products or services; others are intended to enable future architectures. Keeping the current offering and the intended long-term use separate prevents a familiar customer name from being mistaken for evidence of a future product’s completion.
IonQ derives revenue from designing, developing, constructing and selling hardware, providing access to quantum computing systems, consulting and supporting customer programs.[1] Its accounting policy distinguishes obligations recognized over time from products recognized when control passes to the customer.[1] The economic model is therefore not equivalent to a software subscription business. A customized system can generate revenue during development, while maintenance, access and other obligations have different recognition patterns and cash schedules.
Cloud distribution broadens access but does not mean every cloud user is a direct IonQ customer. In cloud-provider arrangements, the annual report identifies the cloud provider as the customer and says recognized revenue reflects the amount charged to that provider rather than the provider’s markup to the end user.[1] Analytical implication: cloud availability demonstrates a route to use, not an automatically large direct enterprise relationship. Paid utilization, contract terms and continuing use determine the financial contribution.
The acquired portfolio includes Capella’s satellite imagery business and Vector Atomic’s precision timing capabilities, alongside networking technologies.[1] SkyWater adds semiconductor manufacturing and advanced packaging services and continues serving foundry customers under its own name.[7] In analytical terms, the company now has both internal technology development and supplier relationships with external customers. The latter can create an independent business opportunity but also requires service reliability and customer confidence beyond IonQ’s own roadmap.
IonQ introduced Superion 256 on September 8, 2026 as its sixth-generation computing platform. It reported fabrication of initial integrated 256-qubit QPUs at SkyWater and the trapping of first ions in prototype systems being built at multiple facilities.[8] These are specific milestones. The announcement did not describe hundreds of installed production systems; it described a platform intended to make future production more repeatable and stated that orders were being accepted for 2027 deliveries.[8]
Electronic Qubit Control is central to that proposition. The September release says the approach controls trapped-ion qubits with electronics integrated on the chip rather than laser-based qubit control, using standard semiconductor processes.[8] This is a manufacturing architecture claim, not a claim that every function of a quantum machine has become ordinary classical electronics. The scale-up still has to combine quantum control, device fabrication, calibration, software, error management and customer operation into a dependable system.
The company reported six tapeouts in the first half of 2026, compression of the design cycle from nine months to two, and twelve times more wafer lots over six months than at a previous foundry.[8] These are company-reported development comparisons. They support the argument that access to SkyWater changes iteration speed. They do not independently quantify finished-system yield, delivered cost per useful logical operation or profitability at volume. Those are later commercial measurements with different denominators.
IonQ describes a roadmap from the 256 generation to a 10K generation with CMOS integrated onto the chip, with both generations under development concurrently.[8] It places full fault tolerance in a 2027 laboratory setting and manufacturable commercial realization in 2028 as expectations.[8] The difference between a lab goal and a commercial goal is material. An experiment can demonstrate a mechanism under tightly managed conditions before that mechanism works reliably in a customer facility at an economic cost.
The company’s forecast of more than a 300-fold reduction in cost per qubit across the roadmap is likewise forward-looking.[8] It is not an audited reduction in the cost of a delivered machine. More importantly, cost per physical qubit is not the same as cost per reliable answer. Error-correction overhead, circuit depth, calibration requirements, throughput and support can change the resources needed for a useful result. A cheaper component only creates the intended economic benefit if the complete system performs adequately.
FIU’s September 22 announcement describes a contract for a Superion 256 with installation expected in late 2027, tied to completion of a campus data center.[12] That is a named customer and a stated physical deployment plan. Its technical and financial significance depends on delivery and acceptance. The release does not disclose an order price, so the installation plan cannot be turned into a quantified revenue contribution simply by multiplying a presumed system price by one machine.
The September 23 NVIDIA research-center announcement describes a planned Superion 256 installation connected to a GB200 NVL72 system through NVQLink, with CUDA-Q orchestration, scheduled for 2027.[13] It presents a joint research program in hybrid software and system prototyping.[13] This is different from evidence of a completed installation or a disclosed large purchase order. The analytical opportunity is integration into a demanding classical-computing environment; the remaining test is what the combined workflow accomplishes.
Quantum benchmarks answer different questions. A physical-qubit count measures hardware resources. A gate fidelity measures the reliability of an operation under a specified test. A logical-qubit result measures performance after encoding and error management. An application result addresses a workload. IonQ’s annual report separately defines physical qubits, quantum gates, noise and photonic interconnects.[1] Treating these categories as interchangeable makes progress look more conclusive than the measurement actually permits.
The arXiv qLDPC preprint reports nine error-correcting codes on a trapped-ion device, including a code encoding four logical qubits into eighteen physical qubits.[18] It describes breakeven performance, with some logical lifetimes comparable to or slightly exceeding physical-qubit lifetimes.[18] Those are useful experimental results, not a demonstration of arbitrary large-scale fault-tolerant applications. The paper’s distinction between code families, memory experiments and physical implementation remains essential to understanding the result.
One example in the paper gives a logical lifetime of 3.95 ± 0.68 seconds versus 3.3 ± 0.9 seconds for physical qubits. The memory experiments post-selected away shots with detected leakage. Conversion of leakage into erasures without that rejection was evaluated in simulations rather than demonstrated in those hardware experiments.[18] The uncertainty ranges overlap. The appropriate description is the paper’s cautious breakeven or marginally exceeding language, not a categorical claim that every logical qubit has become much more reliable. A stronger fault-tolerance case would involve sustained error suppression while more operations, larger systems and more demanding circuits are introduced.
IonQ’s September 22 release describes a real-time decoder evaluated across benchmark circuits simulating up to 408 logical qubits and more than 31.5 million quantum operations, running on a standard CPU.[14] The word simulating matters: this is evidence about a classical decoding architecture and its timing under the studied conditions, not a claim that IonQ operated a physical machine with 408 fault-tolerant logical qubits.[14] It addresses a bottleneck that could otherwise make a quantum processor wait for classical error-processing work.
The October 9 announcement reports more than one thousand entanglement events per second between a trapped-ion qubit and a silicon-vacancy quantum memory.[9] The associated arXiv preprint gives an average rate of 1.03(1) kHz and fidelity of 87.9(7)% for the studied ion-memory Bell pairs, including conversion losses.[17] This is a heterogeneous link, not automatically a complete connection between two large commercial processors. Rate and fidelity need to be read together because faster generation is useful only if the resulting states meet the application’s requirements.
The preprint discusses estimated requirements of ten to one hundred kHz for continuous logical operations between modules under known algorithms and plausible near-term assumptions.[17] That context qualifies the release’s broader distributed-computing language. A one-kHz demonstration is an advance against a prior link benchmark, but scaling channels, improving fidelity and integrating processors remain separate challenges. The measured interconnect result is not proof of an operational quantum data center.
The acquisition sequence is a major part of IonQ’s changing reporting perimeter. The annual report records completed purchases of ID Quantique, Lightsynq, a market-intelligence business, Capella, Oxford Ionics and Vector Atomic during 2025.[1] The June filing adds completed 2026 purchases of Skyloom, Seed Innovations and Nexus Photonics and records SkyWater as a subsequent event.[2] Elisity is different: as of October 13, 2026, it was an announced acquisition subject to closing conditions, not a closed subsidiary.[5][6]
Oxford Ionics closed on September 16, 2025. The annual report gives purchase consideration of approximately $1.589670 billion, comprising $10 million cash and the closing-date fair value of 25,372,150 IonQ shares, including shares withheld for employee taxes.[1] The accounting value at closing is not necessarily the same as a headline valuation stated when a deal was announced. A stock transaction’s accounting consideration can move with the acquirer’s share price between announcement and completion.
The annual report attributes approximately $1.2619 billion of goodwill to Oxford’s workforce and expected synergies and identifies developed technology as the principal acquired intangible asset.[1] It says Oxford’s revenue from acquisition through year-end 2025 was not material.[1] Analytical implication: this was principally an investment in capabilities and future computing architecture rather than the purchase of a large existing revenue stream. Its success is measured by the technical and commercial output of those capabilities over time.
Capella closed on July 11, 2025 for approximately $424.847 million of consideration, including $48.349 million cash and 9,004,982 IonQ shares.[1] Its post-acquisition revenue through December 31, 2025 was $21.5 million.[1] Vector Atomic closed on October 2, 2025 for about $181.5 million of stock consideration, represented by 2,607,452 shares, and contributed $8.7 million of post-acquisition 2025 revenue.[1] These activities have existing customers and different revenue drivers from the original computing program.
Lightsynq closed on May 30, 2025, with purchase consideration of approximately $306.212 million including cash, stock and equity awards; ID Quantique’s controlling stake closed on April 30, 2025 for approximately $116.217 million.[1] IonQ initially acquired about 86% of IDQ and increased ownership to about 91% in July 2025.[1] IDQ contributed $18.8 million of post-acquisition 2025 revenue, while Lightsynq’s reported contribution was not material.[1] Consolidating a controlling stake includes a noncontrolling-interest accounting dimension; it is not the same as owning every share of the subsidiary.
Skyloom closed on January 26, 2026, with June-note consideration of approximately $188.474 million including cash, stock and contingent consideration.[2] Seed closed on January 30, 2026 for approximately $30.076 million of stock consideration and contributed $14 million of revenue from acquisition through June 30.[2] Nexus closed on June 30, 2026 for approximately $76.045 million including stock and equity awards; its contribution through that date was not material.[2] An acquisition at quarter-end can materially change assets and future operations without adding meaningful revenue to that quarter.
SkyWater closed on July 31, 2026, and the closing release specifies $15 in cash plus 0.4883 IonQ shares for each SkyWater share.[7] The June filing’s subsequent-event note records approximately $1.8 billion of total consideration, including 24,135,775 IonQ shares and $741.3 million cash consideration.[2] The exchange ratio was fixed at closing; October movements in IonQ’s share price do not change that completed exchange.
The Elisity agreement was signed on October 9, announced on October 12 and disclosed in an 8-K filed October 13, 2026.[5][6] Closing consideration is $1.15 billion in IonQ shares, subject to specified adjustments, with up to $150 million additional stock consideration tied to revenue milestones from August 1, 2026 through January 31, 2028.[5] The release expects closing in the fourth quarter of 2026, subject to regulatory approval and other conditions.[6] An expected quarter is not a guaranteed closing date.
The share calculation uses a twenty-trading-day VWAP measured before closing, with a price floor of $33.07 and ceiling of $49.61.[5] Separate from merger consideration, IonQ agreed to a $100 million retention pool of RSUs for continuing employees, granted after closing and vesting over four years.[5] The retention pool is additional to the maximum $1.3 billion of merger consideration. It is an additional potential equity cost with a different purpose and timing from the consideration paid to sellers.
At the stated collar endpoints, the unadjusted $1.15 billion closing amount implies about 23.18 million to 34.77 million shares; the maximum $150 million earnout implies about 3.02 million to 4.54 million more, and the $100 million retention pool about 2.02 million to 3.02 million RSUs.[5] These are calculated endpoint illustrations, not the final issued amounts. Leakage adjustments, the closing VWAP, earnout performance and employee vesting affect the actual outcome. Comparing those illustrative amounts with today’s common shares is not a complete pro-forma capitalization.
IonQ reported 2025 revenue of $130.016 million, versus $43.073 million in 2024.[1] Second-quarter 2026 revenue was $80.050 million versus $20.694 million a year earlier, while first-half revenue was $144.718 million versus $28.260 million.[2] Calculated year-on-year growth was approximately 286.8% for Q2 and 412.1% for the half year. These are consolidated GAAP revenue comparisons; the expanding acquisition perimeter prevents them from being treated as pure growth of the original computing business.
The June filing divides Q2 revenue into $46.467 million of quantum hardware and $33.583 million of platform, consulting and support services.[2] For the first half, those categories were $82.176 million and $62.542 million respectively.[2] These reporting categories are not a standalone margin disclosure for each strategic division. In particular, consolidated services can include activities whose economics differ materially from cloud access to a quantum processor. A product label does not show recurring subscription quality by itself.
The August results release defines organic revenue as revenue from IonQ’s quantum-computing products and acquisitions before December 31, 2024.[3] It maintains a 100% full-year organic growth expectation.[3] This is a company-defined supplemental measure and a forecast, not a statement that Q2’s 287% consolidated growth was organic. Later acquisitions fall outside that definition even when their capabilities are useful to the platform. The definition needs to stay stable for later organic comparisons to be meaningful.
The release says approximately 50% of Q2 revenue was international, 60% commercial and 25% multiproduct.[3] These are overlapping descriptions, not slices of one pie that should sum to one hundred percent. A commercial customer can also be international and purchase several products. The company’s commercial definition includes non-U.S.-government enterprise agreements and leading universities.[3] It is broader than a narrow definition of private-sector commercial sales and should not be used to infer that all university or foreign-government exposure has disappeared.
On August 5, IonQ raised 2026 revenue expectations to $280 million to $290 million and explicitly excluded SkyWater contributions.[3] On September 8, it issued an updated $450 million to $460 million range including SkyWater from July 31 through December 31 and eliminating estimated intercompany revenue under their pre-existing commercial agreement.[4] The midpoint increased from $285 million to $455 million, a calculated $170 million difference. That is a bridge between two differently scoped outlooks, not an independently disclosed exact revenue forecast for SkyWater.
IonQ’s accounting policy allocates transaction price to performance obligations and recognizes revenue as obligations are satisfied, using over-time or point-in-time methods as applicable.[1][2] Cash billing may occur on a different schedule. Revenue from work performed is not identical to a new order announced, a deposit collected or the total lifetime value of a customer relationship. In a business with customized hardware and long-cycle contracts, those differences can make quarterly revenue and quarterly cash flow move in different directions.
The June balance sheet reported $105.909 million of net accounts receivable and $87.007 million of total unearned revenue, combining current and noncurrent portions.[2] Receivables represent claims to payment; unearned revenue represents obligations associated with advance payments or recorded billings before performance is complete. Neither can be substituted for RPO. The analytical question is how these balances move as contracts progress: growing receivables without collection would have a different cash implication from advance cash received before delivery.
Q2 2026 operating loss was $337.243 million. Net loss attributable to IonQ was $1.867742 billion, or $5.08 per basic and diluted share.[2] The difference is heavily influenced by a $1.649115 billion loss from changes in warrant-liability fair value.[2] That mark-to-market expense is not a cash payment of the same amount during the quarter. It reflects accounting for instruments whose value changes with market inputs, including IonQ’s share price and volatility assumptions.
This does not make GAAP irrelevant. Warrants are real contractual instruments and can create dilution or other economic consequences. It means the income statement and cash-flow statement answer different questions. A very large fair-value loss can make the bottom line look much worse than contemporaneous operating cash consumption; a fair-value gain can make earnings look much better without making the operating business self-funding. Neither direction should be mistaken for a sudden change in product profitability.
Q2 cost of revenue was $60.110 million, excluding depreciation and amortization, against $80.050 million revenue.[2] The calculated difference is $19.940 million, or approximately 24.9% of revenue, before the separately reported depreciation and amortization line. Without an allocation of depreciation and amortization, that spread is not a fully loaded consolidated gross margin. A cost-of-revenue measure that excludes depreciation is not directly comparable with a peer whose cost of sales includes all relevant depreciation.
Research and development expense was $160.627 million in Q2, sales and marketing $32.895 million, general and administrative expense $117.574 million, and depreciation and amortization $46.087 million.[2] The operating cost base therefore extends well beyond producing the quarter’s recognized sales. In analytical terms, it funds development and a broad corporate platform ahead of the revenue level needed to absorb those costs. Revenue growth will improve operating leverage only if incremental sales contribute more than the additional expenses required to deliver them.
Stock-based compensation included in the Q2 statements was $141.845 million and $270.362 million in the first half, net of capitalized amounts.[2] It is noncash in the current period, but not costless to owners. Awards can transfer part of future company value to employees and other recipients. The useful analysis keeps its cash-flow treatment separate from its per-share consequence, rather than subtracting it and then describing the resulting metric as the full economic cost of operations.
The August release reported Q2 adjusted EBITDA loss of $120.275 million and adjusted EPS of negative $0.33.[3] Its reconciliation excludes specified items including stock compensation, fair-value changes, acquisition and integration costs, depreciation and amortization and other defined adjustments.[3] The non-GAAP measure can help compare aspects of operations, but it is not cash flow and is not interchangeable with operating loss. Different businesses may exclude different items under the same adjusted label.
First-half 2026 net cash used in operating activities was $254.781 million, versus $85.599 million in the first half of 2025.[2] Equipment purchases were $18.583 million in H1 2026.[2] Their calculated sum, $273.364 million, is a historical operating-plus-equipment consumption measure, averaging about $45.561 million per month over six months. It excludes acquisition consideration and strategic investments, so it is not total corporate cash deployment or a forecast of future monthly spending.
At June 30, 2026, IonQ reported $1.235729 billion of cash and equivalents, $883.240 million of short-term investments and $840.365 million of long-term investments.[2] Their calculated total is $2.959334 billion. That is the financial asset aggregate underlying the company’s rounded $3.0 billion description.[3] Long-term investments are not the same balance-sheet category as cash and equivalents, even if some securities may be liquid. The aggregate comprises cash and equivalents, short-term investments and long-term investments; it is not cash alone.
The August release gives a rounded $2.0 billion pro-forma cash, equivalents and investments balance after cash consumed in consummating the SkyWater acquisition.[3] Separately, the June filing records $741.3 million of cash purchase consideration for SkyWater.[2] Those disclosures measure different scopes. Subtracting the $741.3 million seller cash consideration from June’s $2.959334 billion aggregate yields approximately $2.218 billion, not the rounded $2.0 billion pro-forma balance. The seller consideration alone therefore does not reconcile the two disclosures.
Dividing June’s $2.959334 billion aggregate by the calculated H1 operating-plus-equipment average of $45.561 million per month gives approximately sixty-five months of arithmetic coverage.[2] This is not a runway forecast. Both numerator and denominator precede the SkyWater closing, the denominator excludes acquisitions and strategic investment purchases, and future foundry operations and technology development can change consumption. A historical ratio can illustrate scale without predicting the date on which financing will become necessary.
June total liabilities were $3.439012 billion, including $3.052398 billion of warrant liabilities.[2] That warrant balance is a fair-value liability, not conventional funded debt or a same-sized bank borrowing. The balance sheet also includes payables, accrued expenses, unearned revenue and leases, which have different economic and payment characteristics.[2] Treating all liabilities as debt would materially distort a liquidity comparison; treating all nonbank obligations as irrelevant would distort it in the opposite direction.
The balance sheet also reported $2.185971 billion goodwill and $778.874 million net intangible assets at June 30.[2] These assets are distinct from liquid resources. Their value depends on acquired technologies, customer relationships and expected benefits. Impairment would be an accounting consequence of weaker expectations; it would not create a new cash outflow equal to the charge at that moment. Nevertheless, it could indicate that earlier capital allocation produced less value than anticipated.
June 30 RPO was approximately $485 million, up from approximately $370 million at December 31, 2025.[2][1] IonQ expected about 50% of June RPO to be recognized as revenue over the following twelve months.[2] That expectation is a timing estimate, not revenue already earned. It is also not a statement that half will necessarily be collected as cash during the same period. Recognition, invoicing and receipt can follow different contractual schedules.
The filing explicitly includes funded orders, for which funding has been authorized and appropriated, and unfunded firm orders, for which funding has not been appropriated.[2] Unexercised contract options are excluded until exercised.[2] The $485 million total is therefore not an all-funded backlog number. Without a separately disclosed funded split, assigning the entire balance the strongest possible funding status would exceed what the filing says.
Bookings generally describe contracting activity over a period, whereas RPO measures unsatisfied obligations at a reporting date and revenue measures performance recognized during a period. They are not interchangeable indicators. A new contract can enter RPO, part of an existing contract can leave RPO through revenue recognition, and an acquisition can add obligations without representing a newly won organic order. A useful commercial bridge explains those movements rather than placing the three labels beside one another as if they measure the same thing.
IonQ’s October 7 release gives the Stage C QBI agreement a potential value of up to $300 million and explicitly says funding beyond the initial obligation remains subject to future appropriations and DARPA funding actions.[10] It describes Stage C running through 2029, with successive Superion generations entering independent verification and validation.[10] The ceiling cannot be counted as $300 million of current funded backlog or revenue. The initial obligation and later funding actions have their own meaning.
DARPA independently identifies IonQ among the teams advancing and explains that QBI evaluates credible paths to utility-scale operation rather than selecting one winning company.[11] The agency’s broader initiative asks whether computational value can exceed cost by 2033.[11] That 2033 program question is not the same as IonQ’s nearer commercial roadmap. External validation is valuable precisely because it tests the assumptions; selection for the process does not show that those assumptions have already passed every hardware test.
The corrected August 6 clock release states that a $28 million contract modification covers manufacturing development and twenty-five Evergreen-05 clocks.[15] A further $30 million option, not yet exercised, would cover one hundred additional clocks, giving a potential combined value of up to $58 million.[15] Up to 125 clocks is therefore a possible combined scope, not a claim that every unit is already funded, delivered or recognized in sales. The distinction survives even when the product itself has an established technical history.
The February 23 SHIELD announcement describes a $151 billion IDIQ ceiling across a framework in which more than 2,400 companies were eligible to compete for future task orders.[16] The ceiling is program-wide, not an IonQ award of that amount. Framework participation is access to potential procurement; a funded task order would be a later, narrower fact. The program-wide ceiling is neither IonQ’s RPO nor its annual revenue.
The August results release separately lists memoranda of understanding with Anduril and Sandia.[3] A memorandum provides a collaboration framework; it is not automatically a purchase order with a disclosed value. The key evidence chain is specific contracted work, funded obligations, completed performance and cash collection. Government exposure can support long-duration development, but appropriations, competitive selection and program priorities remain independent risks at each step.
IonQ’s common shares outstanding rose from 221,919,191 at December 31, 2024 to 362,592,722 at December 31, 2025.[1] The annual statement of changes in equity separates acquisitions, equity issuance, compensation-related shares and warrant exercises.[1] The rise is not attributable to one event. It shows why changes in total company resources and changes in the ownership denominator need to be analyzed together. A larger cash reserve was achieved alongside a much larger number of shares.
The July 2025 offering closed with approximately $977.2 million net proceeds, and the October 2025 offering with approximately $1.9771 billion net proceeds.[1] Both included common shares or prefunded warrants and additional private warrants.[1] Prefunded warrants, once exercised, become part of issued ownership and must not be added again as if still outstanding. Ordinary private warrants remain a different category of potential future shares and have much higher exercise prices.
At June 30, 2026, 36,042,530 Series A private warrants remained outstanding with a $99.88 exercise price and 43,010,800 Series B private warrants with a $155 exercise price.[2] The June filing states that the Series A and B prefunded warrants were no longer outstanding.[2] Together, the private warrants represent 79,053,330 potential common shares subject to their terms. Their exercise prices are contractual facts, not author-created target prices for IonQ stock.
The private warrants are liability-classified and remeasured using valuation models.[2] That accounting is responsible for substantial earnings volatility, but the instrument count is separate from the dollar fair value. A $3 billion liability balance does not mean three billion additional shares or a contemporaneous $3 billion cash payment. Conversely, a lower fair value at a later reporting date would not extinguish the contractual warrants unless their terms or status changed.
The June filing reported 1,065,043 public warrants with an $11.50 exercise price and no redemptions as of June 30.[2] The August 28 filing subsequently specified expiration on September 30, 2026 and cessation of IONQ WS trading before the September 29 market open.[20] NYSE then filed its September 30 Form 25 notice of removal for that public-warrant class.[21] The expiration date has passed. The June public-warrant count is historical, not a current dilution balance; it must not be carried into the October capital structure as though those warrants still remained exercisable. The high-strike private Series A and B warrants have separate terms and are not that expired public series.
June outstanding RSUs were 17,046,325; options were 1,836,415; restricted stock was 8,957,744.[2] The filing separately reported 3,954,182 PSUs outstanding and 8,199,699 expected to vest under its performance assessment, with additional qualification about awards whose grant date for accounting purposes had not yet been set.[2] Those categories cannot simply be summed into one present outstanding-common figure. Grant, expected vesting, vesting, settlement and issued common stock are different stages.
The Elisity retention pool would add another layer of equity awards if the acquisition closes.[5] Its four-year vesting schedule differs from the seller consideration and earnout.[5] Analytical implication: preserving cash through stock consideration does not remove the need for discipline. The acquired business must generate enough value to justify the sellers’ ownership, contingent payments and incentives required to retain employees. The long-term commercial outcome and the long-term per-share outcome can diverge.
The reference closing price is $34.73 on October 13, 2026.[19] Finviz’s same-date market fields give approximately $14.07045 billion of market capitalization and 405.14 million shares outstanding.[19] Multiplying the displayed price by the displayed share count produces approximately $14.070512 billion, effectively matching the displayed capitalization within rounding. This is a dated market-provider comparison, not an audited company capitalization table.
The June filing’s cover gives 381,002,314 common shares outstanding at July 29, 2026, before the SkyWater closing.[2] SkyWater then involved 24,135,775 IonQ shares according to the subsequent-event note.[2] Those historical dates help explain why the later provider denominator is larger, but their sum is not a certified exact October share count. Compensation settlements, other issuance and the provider’s measurement conventions can affect the denominator. The dated company figures and the dated market fields retain their distinct provenance.
Dividing the approximate October market capitalization by the September 2026 revenue-guidance midpoint of $455 million yields a calculated ratio of about 30.9 times forecast annual revenue.[19][4] It uses a company forecast rather than an achieved result and it is an equity-value-to-sales ratio, not enterprise value to sales. The guidance also includes SkyWater for only the stated post-closing portion of 2026. The ratio does not value a full normalized year of the combined business.
The ratio illustrates substantial expectations but does not show whether shares are expensive or inexpensive. Revenue quality, future margins, cash needs and the value of technologies still under development affect the economic interpretation. Comparing a foundry-inclusive forecast with a pure software or pure computing company without adjusting for business mix can be misleading. A consistent comparison would use the same valuation measure, period, acquisition perimeter and accounting definition.
Finviz’s October 13 fields report a 399.03 million share float, 48.97% institutional ownership, 1.51% insider ownership, 10.18% short float and a 1.90 short ratio.[19] These are vendor measures with underlying reporting and calculation conventions; October 13 is the market reference date, not a claim that every beneficial-owner filing or short position was newly reported that day. Aggregate institutional ownership does not identify an active investor’s current conviction or a passive fund’s reason for holding the security.
IonQ’s strategy asks a broad collection of technologies to become a coordinated commercial platform. The annual report warns about integrating acquired operations, retaining personnel, managing new capabilities and realizing anticipated benefits.[1] Each transaction can add talent and customers while also adding systems, controls and obligations. A larger product portfolio creates cross-selling possibilities, but it can also make management attention and accountability harder to allocate.
Oxford’s electronic-control technology and SkyWater’s manufacturing access are a clear example of a proposed technical fit. The September product announcement links those acquisitions to Superion development.[8] The integration still has to work in finished systems. Semiconductor yields, packaging, hardware reliability and the qubit-control architecture are interdependent; success in one does not neutralize failure in another. A fast design cycle can shorten learning while still requiring multiple iterations before a commercial device is dependable.
On October 13, 2026, SkyWater and D-Wave announced a multiyear agreement extending specialized semiconductor process development and manufacturing at SkyWater’s Minnesota fab, with terms protecting D-Wave’s intellectual property.[22] This is a concrete example of the subsidiary serving an external quantum-computing customer even after its acquisition by IonQ. The release does not disclose a contract value, so it cannot show a quantified addition to IonQ revenue or backlog. D-Wave’s processor performance belongs to D-Wave, not to IonQ’s trapped-ion computing results.
DARPA’s Stage C group includes neutral-atom, silicon-spin, superconducting and trapped-ion approaches, alongside teams from the precursor program.[11] That diversity shows why no single headline metric settles the competitive question. Each modality has trade-offs involving control, fidelity, connectivity, manufacturing, speed and error-correction overhead. Commercial competition also includes what classical computing can accomplish at the required accuracy and cost, not just which quantum machine has the highest physical-qubit count.
Security carries its own technical and commercial risks. Elisity’s release describes combining present identity-based controls with cryptographic inventory and quantum-related protection.[6] That is a proposed integrated roadmap, not an already delivered universal security solution. Customers may require interoperability, compliance, deployment support and evidence that new products do not disrupt existing networks. Cross-selling a sophisticated security product is not equivalent to attaching a new label to an existing customer relationship.
The annual auditor identified developed-technology valuation in the Capella and Oxford acquisitions as a critical audit matter because of significant judgment and assumptions about future results.[1] This is not an adverse audit opinion; it identifies a difficult area of estimation. The distinction matters. Acquired technology can be real and valuable while its purchase-price allocation depends on uncertain growth and profitability assumptions. Later operating evidence is what tests those assumptions.
Customer concentration can amplify quarterly variability. The annual report records three significant customers accounting for 53% of 2025 revenue.[1] A broader acquired base may diversify activity, but one completed milestone, delivery schedule or large customer can still affect a reporting period materially. Repeat business, payment performance and customer diversification provide a more useful assessment of durability than a long list of possible markets.
The nearest announced corporate window is the expected fourth-quarter 2026 Elisity closing, subject to approvals and customary conditions.[6] The agreement’s January 7, 2027 outside-date termination provision is a legal condition, not a promised completion date.[5] The useful update would identify whether closing occurred, the actual share consideration and the treatment of the retention pool. Until completion is announced, Elisity remains outside the closed-acquisition list.
Superion 256 customer deliveries are expected in 2027; FIU’s installation is expected in late 2027 and the NVIDIA research-center installation is scheduled for 2027.[8][12][13] These are separate deployments with their own conditions. A broad delivery year does not give the exact date of each acceptance event. Confirmed completion would be a stronger fact than another statement of intended timing.
The Superion roadmap’s 2027 laboratory and 2028 commercial fault-tolerance expectations are company targets, while DARPA Stage C runs through 2029 under the company’s announcement.[8][10] DARPA’s wider utility-scale inquiry uses a 2033 objective.[11] The clocks measure different programs and should remain different. A technical ambition can change without changing a legal closing deadline; a government evaluation horizon is not a substitute for a customer delivery commitment.
The commercial watchlist includes organic growth under the company’s stated definition, the movement of RPO into revenue, receivable collection and the economics of multiproduct relationships.[3][2] A larger acquired business can lift reported revenue immediately; stronger customer adoption should also become visible through renewals, additional purchases or repeat system deployments. Acquisition-driven revenue growth and stronger customer adoption have different strategic implications.
The status and evidentiary weight differ across these layers. Closed transactions are closed, announced transactions remain conditional, recognized revenue belongs to its reporting period, and technical demonstrations establish only the measured result. Future roadmaps and valuation ratios are tools for asking questions, not facts about an already achieved future. That distinction leaves room for genuine progress without converting ambition into a promise or risk analysis into an investment recommendation.
No. It closed on July 31, 2026, with $15 cash and 0.4883 IonQ shares per SkyWater share.[7] It is outside Q2 operating results because closing occurred after June 30.[2]
As of October 13, 2026, Elisity remained a signed transaction expected to close in Q4, subject to approvals and conditions.[5][6] Its retention pool is additional to merger consideration.[5]
No. Q2 consolidated revenue grew approximately 287%; the company defines organic revenue separately and expects 100% full-year organic growth.[3] Later acquisitions are outside that supplemental definition.
No. Funding beyond the initial obligation depends on future appropriations and funding actions.[10] Stage C is independent evaluation of the path to useful systems, not certification that utility scale has already been reached.[11]
The September announcements describe orders and planned installations: initial customer deliveries in 2027, FIU installation in late 2027, and NVIDIA research-center installation in 2027.[8][12][13] These remain announced schedules.
Q2 net loss attributable to IonQ was $1.867742 billion, including a $1.649115 billion warrant fair-value loss.[2] H1 operating cash use was $254.781 million.[2] The periods and measurement bases differ; a fair-value charge is not the same amount of cash consumed.
It demonstrates a particular ion-memory link at 1.03(1) kHz and 87.9(7)% fidelity.[17] Scalable processors, multiple channels, reliability and useful distributed applications are additional technical tests.
No. It is an explicitly editorial description of financial and operating robustness. It is not a buy, sell or hold instruction, a price target or an assessment of whether the current share price is appropriate.
Every Merlintrader stock hub, catalyst update and market brief is published to Telegram the moment it goes live. No paywall, no spam, just the research.
Disclaimer. This content is published by Merlintrader for educational and informational purposes only. It is independent journalism and research. It does not constitute investment advice, an investment recommendation, an offer or a solicitation to buy or sell any security, and it is not a research report within the meaning of applicable United States securities regulation. Nothing here should be read as a recommendation to buy, sell or hold $IONQ or any other security.
Figures are taken from public filings with the U.S. Securities and Exchange Commission, company press releases and market-data providers, and are stated with their reference dates. Data can change without notice, and figures published before a results release become outdated the moment that release is issued. Merlintrader makes no representation that the information is complete or current at the time of reading. Readers should verify every figure against the primary source before acting on it.
Defense, space and technology companies carry procurement, execution, regulatory and financing risks. Government orders can be delayed or cancelled, production can exceed budget, export restrictions can limit sales, and acquisitions may fail to produce the expected returns. Contract announcements do not guarantee profitable deliveries or cash collection. Additional borrowing or equity issuance can increase obligations or dilute existing shareholders. Investors can lose part or all of their capital. Every reader is responsible for their own decisions and should consult a licensed financial adviser where appropriate.
Merlintrader may hold positions in securities mentioned. Some links on this page are affiliate or referral links, including those to Finviz and Stocktwits, which may generate a commission at no cost to the reader. Full legal information is available on the disclaimer and terms of use and privacy pages.
Conferences, industry summits, government-facing events and company watchpoints across the sector, in one free calendar.
Open the Space, Defense & AI Event Calendar →
