Adam Brown 9/3/26 Adam Brown 9/3/26

QumulusAI Completes Deployment of All 616 NVIDIA RTX PRO 6000 Blackwell GPUs for Runpod, Weeks Ahead of Schedule

Full activation across 77 GPU nodes — combining one- and two-year reserved-capacity agreements — was reached in August 2026, ahead of the original Sept. 1, 2026 target

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced that it has completed deployment of 616 NVIDIA RTX PRO 6000 Blackwell GPUs for Runpod under two reserved-capacity agreements combining one- and two-year terms.

All 77 GPU nodes under the agreements are now active. Full deployment was reached in August 2026, ahead of the agreements' original Sept. 1, 2026 target. The GPUs are being served from QumulusAI's existing U.S. data center footprint.

"Teams building on Runpod need inference capacity that lands fast and prices well. QumulusAI got all 616 of these GPUs live ahead of schedule, which means developers get to use them sooner," said Bill Sehmel, Manager of Datacenter Infrastructure at Runpod.

Runpod, a GPU cloud marketplace for AI developers, will offer the RTX PRO 6000 Blackwell capacity to its customers for production AI inference and agentic workloads. RTX PRO 6000 Blackwell GPUs extend QumulusAI's inference- and visualization-class compute alongside its NVIDIA Blackwell B300 and B200 fleet, consistent with the company's approach of matching each workload to the right architecture.

"Bringing all 616 of these RTX PRO 6000 GPUs live for Runpod weeks ahead of schedule is exactly what our demand-led model is built to do," said Ryan DiRocco, CTO of QumulusAI.

About QumulusAI

QumulusAI is a distributed AI cloud platform that delivers accelerated access to high-performance GPU compute. Through an inference-first, demand-led deployment model across a network of data center sites, QumulusAI brings compute closer to customer demand, helping AI teams and enterprises scale production AI workloads with speed, flexibility and control. By combining rapid deployment with flexible private cloud infrastructure, QumulusAI gives customers a faster, more adaptable path beyond the capacity constraints of traditional centralized and hyperscale cloud models. Learn more at QumulusAI.com.

About Runpod

Runpod is the AI developer cloud. The platform provides the infrastructure AI developers need across the full lifecycle: experiment, train, fine-tune, deploy and scale. Over 1 million developers build on Runpod. Specifically for AI workloads, Runpod is the fastest path from AI experiment to production. For more information, visit runpod.io.

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company’s reserved-capacity agreement with Runpod for NVIDIA RTX PRO 6000 Blackwell GPUs, the anticipated term of the agreement, the company’s ability to serve the agreement from its existing data center footprint, and Runpod’s offer and use of the capacity to serve its customers for production AI inference and agentic workloads. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), and the company’s quarterly report on Form 10-Q for the quarter ended June 30, 2026, as amended, as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 8/26/26 Adam Brown 8/26/26

QumulusAI Amends Oklahoma Lease, Extending Potential Tenure Through January 2044 at a Site With 19 MW of Contracted Power

The amendment replaces a single five-year renewal option with three consecutive five-year renewal options, giving the company long-dated control of all 19 MW of contracted power at the location.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced an amendment to the surface lease at its Oklahoma data center, extending the company’s potential tenure at the site through January 31, 2044.

The amendment replaces the single five-year renewal option in the original lease with three consecutive five-year renewal options, each exercisable at the company’s election upon written notice 60 days’ prior to the expiration of the applicable term, assuming continued compliance with the agreement. QumulusAI is not obligated to exercise any of them.

The Oklahoma location, approximately 10.57 acres, has been in service since 2024 and is the company’s legacy operating site. It carries 19 MW of total contracted power capacity. Of that, 10 MW is currently active and supports legacy blockchain mining and hosting operations, and 9 MW is available and unutilized.

“Long-dated control of an established, energized site is a different asset than a lease with a runway,” said Scott Krosnowski, CFO of QumulusAI. “The renewal options here are ours to exercise, and that optionality is what makes the site’s available power worth planning around.”

The amendment also permits leasehold financing at the site, which can support project-level capital structures for future development there.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company’s amendment to the surface lease at its Oklahoma data center and the potential extension of its tenure at the site through January 31, 2044, whether and when the company exercises any of the three five-year renewal options, the company’s continued compliance with the agreement, the anticipated benefits of long-dated site control and the optionality the company expects it to provide, the company’s continued control of the site’s 19 MW of contracted power capacity, the availability of the 9 MW currently unutilized as an expansion path for potential future high-performance computing expansion and the timing of any conversion of that capacity, and the availability of leasehold financing at the site and its anticipated support for project-level capital structures. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contactinvestors@qumulusai.com

Media Contactmedia@qumulusai.com

Adam Brown 8/25/26 Adam Brown 8/25/26

QumulusAI Reports Second Quarter 2026 Results

Revenue more than doubles year over year to $6.7 million as contracted GPU capacity continues to come online; cumulative signed customer contract value reaches $282.5 million.

ATLANTA, Aug. 25, 2026 — QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today reported financial results for the second quarter ended June 30, 2026. This is the company’s first quarterly report since it began trading on the Nasdaq Global Market on July 16, 2026.

"This was the quarter our model started proving itself," said Michael Maniscalco, CEO of QumulusAI. "AI Compute we had already sold came online and started generating revenue. We signed 21 new direct contracts in the quarter, and last week we contracted up to 3.75 MW in metropolitan Atlanta, our home market, with potential to expand at the same site. Demand is not our constraint. Deploying against it faster than competitors is our goal."

"Revenue more than doubled year over year and gross margin expanded to 67%, as GPU activations grew revenue faster than colocation costs," said Scott Krosnowski, CFO of QumulusAI. "Our most recent Blackwell contracts are generating between $18 million and $20 million of annualized revenue per megawatt, compared with just over $16 million across the installed base. Customers are also committing and paying ahead of delivery — deferred revenue rose $30.5 million in the first half, contributing to $22.3 million of operating cash flow."

Second Quarter 2026 Financial Highlights

All comparisons, unless otherwise noted, are to the three months ended June 30, 2025.

Revenue of $6.7 million, an increase of $3.6 million, or 118%, from $3.1 million. Compute power revenue grew to $5.6 million, or approximately 84% of total revenue, from $1.3 million, or approximately 43% of total revenue.
Gross profit of $4.5 million, an increase of $2.8 million, or 163%, from $1.7 million. Gross margin expanded to 67% from 55%.
Operating loss of $7.7 million, compared to $2.2 million. The increase reflects a $5.8 million rise in depreciation and amortization tied to expanded HPC (High-Performance Computing) infrastructure, as well as higher general and administrative costs associated with public company readiness and headcount growth.
Net loss of $22.8 million, compared to net income of $12.1 million. The current period includes a $19.2 million non-cash loss on the issuance of convertible notes; the prior-year period included a $14.5 million non-cash gain on remeasurement of the Company's investment in The Cloud Minders.
Adjusted EBITDA loss of $0.8 million, compared to a loss of $0.3 million, as revenue growth was offset by increased operating costs associated with public company readiness and personnel.

  
    
        ($ in thousands, except per share)
        Q2 2026
        Q1 2026
        Q2 2025
        H1 2026
        H1 2025
      

    
        Revenue
        $6,713
        $3,420
        $3,085
        $10,133
        $4,957
      

        Cost of revenue
        2,242
        2,136
        1,386
        4,378
        2,601
      

        Gross profit
        4,470
        1,284
        1,699
        5,755
        2,355
      

        Gross margin
        66.6%
        37.5%
        55.1%
        56.8%
        47.5%
      

        Operating loss
        (7,671)
        (5,527)
        (2,182)
        (13,197)
        (3,111)
      

        Net income (loss)
        (22,776)
        (49,617)
        12,119
        (72,393)
        10,296
      

        Adjusted EBITDA (non-GAAP)(1)
        (782)
        (2,790)
        (266)
        (3,572)
        (431)
      

  

Figures are rounded to the nearest thousand; totals may not sum due to rounding.

(1) Adjusted EBITDA is a non-GAAP financial measure. See the reconciliation of net income (loss) to Adjusted EBITDA included at the end of this release.

Second Quarter Operational Highlights

Customer and Demand

Signed 21 new direct customer AI compute contracts during the quarter with aggregate expected take-or-pay contract value of $169.7 million.
Direct customer relationships grew to more than 96% of the recurring revenue base at quarter end, from less than 10% a year earlier, as the Company completed its transition away from dependence on a single marketplace.
AI Compute revenue reached 84% of total revenue, up from 61% in the first quarter of 2026 and 43% in the second quarter of 2025.

Infrastructure and Capacity

Grew the deployed GPU fleet from 952 to 3,088, an increase of approximately 224%.
Ended the quarter with 8 MW of HPC capacity under executed lease and colocation agreements.

Recent Corporate Developments

Trading on the Nasdaq Global Market began under the ticker symbol "QMLS" on July 16, 2026.
Became an NVIDIA Cloud Partner on July 17, 2026.
Signed more than $120 million in new customer agreements, including a three-year agreement valued at more than $71 million.
Signed a GPU-as-a-Service agreement with DRW, a diversified trading firm innovating across both traditional and cutting-edge markets.
Purchased 1,632 NVIDIA Blackwell B300 GPUs to meet accelerating customer demand.
Entered a colocation agreement in metropolitan Atlanta for up to 3.75 MW, with a right of first offer on up to 7 MW of expansion capacity at the same site.

Webcast and Conference Call

QumulusAI will host a conference call and webcast today at 5:00 p.m. Eastern Time to discuss second quarter 2026 results. The live webcast, together with this release and supplemental materials, is available at investors.qumulusai.com. A replay will be available on the same site following the call.

Non-GAAP Financial Measures

To supplement its condensed consolidated financial statements, which are prepared and presented in accordance with generally accepted accounting principles in the United States (“GAAP”), the company presents Adjusted EBITDA, a non-GAAP financial measure. QumulusAI’s reasons for use of this measure and reconciliation of this non-GAAP financial measure to the most directly comparable GAAP measure and other information are included at the end of this release. Adjusted EBITDA has limitations as an analytical tool and should not be considered in isolation or as a substitute for net income (loss) or any other measure prepared in accordance with GAAP.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company’s continued ability to bring contracted GPU capacity online; the amount and timing of revenue the company expects to recognize from its remaining performance obligations and its signed customer contracts; the company’s ability to deploy against demand faster than competitors; the company’s annualized revenue per-megawatt and realization thereof; the receipt of customer payments ahead of delivery as committed; the company’s plans to expand energized capacity, including under the right of first offer at its metropolitan Atlanta site; and the company’s expected finance lease payments. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), and the company’s quarterly report on Form 10-Q for the quarter ended June 30, 2026, as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Condensed Consolidated Statements of Operations (Unaudited)

	Three Months Ended June 30,		Six Months Ended June 30,
	2026	2025	2026	2025
Revenue
Revenue from cryptocurrency mining	$410,081	$148,038	$779,060	$295,839
Revenue from mining hosting services	693,778	1,625,701	1,642,205	3,349,152
Revenue from compute power	5,608,946	1,311,700	7,711,458	1,311,700
Total revenue	6,712,805	3,085,439	10,132,723	4,956,691
Costs and expenses
Cost of revenue	2,242,402	1,386,374	4,378,199	2,601,468
General and administrative expenses	4,342,620	2,454,605	8,250,275	3,263,676
Sales and marketing expenses	916,220	366,093	1,152,572	664,293
Depreciation and amortization expense	6,882,154	1,059,900	9,548,999	1,537,948
Total costs and expenses	14,383,396	5,266,972	23,330,045	8,067,385

Operating loss	(7,670,591)	(2,181,533)	(13,197,322)	(3,110,694)
Other income (expenses)
Income from equity method investments	—	629,816	21,994	864,320
Gain on sale of equity method investments	—	—	12,569,661	—
Gain on remeasurement of investment in TCM	—	14,549,536	—	14,549,536
Change in fair value of warrant liability	—	—	(1,585,838)	(692,103)
Change in fair value of digital assets	—	85,756	—	37,984
Change in fair value of convertible note	2,380,000	—	2,380,000	—
Change in fair value of additional convertible notes option	3,850,850	—	3,850,850	—
Gain on sale of property and equipment	1,034	—	36,298	—
Loss on issuance of convertible note	(19,241,000)	—	(73,881,850)	—
Loss on extinguishment of debt	—	(71,094)	—	(153,834)
Other income (expense), net	(113,905)	(7,527)	(187,750)	26,954
Interest expense, net	(2,045,748)	(601,260)	(2,584,024)	(940,946)
Total other income (expenses), net	(15,168,769)	14,585,227	(59,380,659)	13,691,911

Income (loss) before income tax expense	(22,839,360)	12,403,694	(72,577,981)	10,581,217
Income tax expense (benefit)	(63,154)	285,120	(185,064)	285,120
Net income (loss)	$(22,776,206)	$12,118,574	$(72,392,917)	$10,296,097
Net income (loss) in non-controlling interests	100,128	—	(150,711)	—
Net income (loss) attributable to common shareholders	$(22,876,334)	$12,118,574	$(72,242,206)	$10,296,097

Net income (loss) per share, basic	$(0.72)	$0.71	$(2.28)	$0.66
Net income (loss) per share, diluted	$(0.72)	$0.46	$(2.28)	$0.43
Weighted-average common stock outstanding, basic	31,740,634	16,983,356	31,680,098	15,500,358
Weighted-average common stock outstanding, diluted	31,740,634	26,486,792	31,680,098	24,239,377

Condensed Consolidated Balance Sheets

	June 30, 2026 (unaudited)	December 31, 2025
ASSETS
Current assets:
Cash	$19,967,188	$11,712,493
Restricted cash	19,925,104	—
Accounts receivable, net of allowance for credit losses of $365,133 and $2,263 as of June 30, 2026 and December 31, 2025, respectively	11,058,423	57,889
Prepaid expenses and other current assets	2,253,755	1,134,851
Total current assets	53,204,470	12,905,233
Property and equipment, net	44,006,002	12,502,886
Operating right-of-use assets, net	1,374,332	1,438,970
Finance right-of-use assets, net	47,919,004	6,996,077
Equity method investments	—	4,227,130
Investment in equity securities	1,000,000	—
Deposits on power equipment	26,022,880	13,622,641
Goodwill	31,416,827	31,416,827
Intangible assets, net	7,227,864	7,268,513
Other assets	2,830,837	1,356,216
Total assets	$215,002,216	$91,734,493
LIABILITIES AND SHAREHOLDERS' EQUITY (DEFICIT)
Current liabilities:
Accounts payable	$7,546,971	$1,248,175
Dividend payable	359,188	359,188
Accrued expenses and other current liabilities	4,676,188	2,833,337
Deferred revenue	30,460,939	—
Current portion of notes payable	1,158,583	1,684,554
Current portion of notes payable – related party	2,000,000	3,848,915
Current portion of USD.AI protocol loans	6,892,685	—
Operating lease liabilities – current portion	106,795	97,463
Finance lease liabilities – current portion	13,067,517	1,645,069
Deferred tax liability	238,317	423,381
Total current liabilities	66,507,183	12,140,082
Long-term notes payable, net of current portion	5,917,155	6,241,948
Operating lease liabilities	1,459,005	1,497,549
Finance lease liabilities	32,708,652	5,179,828
Warrant liability	2,968,793	1,382,955
Additional convertible notes option liability	38,721,000	—
USD.AI protocol loans, net of current portion	12,020,692	—
Convertible note payable	55,481,000	—
Total long-term liabilities	149,276,297	14,302,280
Total liabilities	215,783,480	26,442,362
Commitments and contingencies (Note 24)
Shareholders' Equity (Deficit)
Common stock – no par value; 500,000,000 shares authorized, 31,727,001 and 31,367,559 shares issued and outstanding as of June 30, 2026 and December 31, 2025, respectively	99,393,515	93,400,180
Additional paid-in capital	6,827,270	6,318,290
Accumulated deficit	(109,788,460)	(37,546,254)
Total shareholders' equity (deficit) attributable to QumulusAI shareholders	(3,567,675)	62,172,216
Non-controlling interests	2,786,411	3,119,915
Total shareholders' equity (deficit)	(781,264)	65,292,131
Total liabilities and shareholders' equity (deficit)	$215,002,216	$91,734,493

Condensed Consolidated Statements of Cash Flows (Unaudited)

	For the Six Months Ended June 30,
	2026	2025
CASH FLOWS FROM OPERATING ACTIVITIES:
Net loss	$(72,392,917)	$10,296,097
Adjustments to reconcile net income (loss) to net cash used in operating activities:
Depreciation and amortization expense	4,967,754	1,537,948
Amortization of loan origination costs	33,883	17,187
Amortization of discount on convertible note	—	135,334
Bad debt expense	302,244	—
Amortization of premium on loan receivable	—	(16,281)
Non-cash interest expense	—	6,418
Recovery of credit losses	—	(36,921)
Amortization of right-of-use assets	4,645,883	462,598
Interest expense under finance lease obligations	1,756,748	133,297
Income from equity method investments	(21,994)	(864,320)
Gain on sale of equity method investments	(12,569,661)	—
Gain on remeasurement of investment in TCM	—	(14,549,536)
Change in fair value of warrant liability	1,585,838	692,103
Change in fair value of digital assets	—	(37,984)
Change in fair value of convertible note	(2,380,000)	—
Change in fair value of additional convertible notes option	(3,850,850)	—
Change in deferred taxes	(185,064)	285,120
Stock-based compensation	241,809	250,552
Issuance of warrants for services	74,659	—
Issuance of warrants as consideration payable to customer	192,512	—
Gain on sale of property and equipment	(36,298)	—
Loss on issuance of convertible note	73,881,850	—
Loss on extinguishment of debt	—	153,834
Changes in operating assets and liabilities:
Accounts receivable	(11,302,778)	37,057
Due from related party	—	(1,590)
Prepaid expenses and other current assets	(1,163,945)	(102,410)
Proceeds from sale of digital assets	102,068	1,438,271
Deposits	—	69,672
Mining of digital assets	(102,068)	(1,290,847)
Accounts payable	6,298,796	1,158,630
Accrued expenses	1,842,851	11,192
Deferred revenue	30,460,939	—
Operating lease liabilities	(29,212)	(30,219)
Intangible assets	(47,833)	(3,600)
Due to related party	—	(547,484)
Net cash provided by (used in) operating activities	22,305,214	(795,882)
CASH FLOWS FROM INVESTING ACTIVITIES:
Purchase of property and equipment	(36,824,254)	(1,014,539)
Proceeds from sale of property and equipment	478,164	—
Proceeds from collections of loans receivable	—	285,654
Deposits on mining equipment	—	(313,088)
Deposits on power equipment	(14,146,100)	—
Data center set up costs	(1,474,621)	—
Purchase of equity securities	(1,000,000)	—
Proceeds from sale of U.S. dollar coin	—	391,584
Cash acquired as part of business acquisition	—	2,449,042
Proceeds from sale of equity method investments	16,559,622	—
Distributions from equity method investments	259,163	1,904,000
Distributions to joint venture partners	(182,793)	—
Net cash (used in) provided by investing activities	(36,330,819)	3,702,653
CASH FLOWS FROM FINANCING ACTIVITIES:
Proceeds from sale of common stock, net of issuance costs	5,980,834	1,897,028
Repayments on finance lease obligations	(6,518,746)	(521,865)
Proceeds from convertible note payable	28,800,000	—
Proceeds from exercise of warrants	12,501	—
Payment of debt issuance costs	(2,249,000)	—
Proceeds from line of credit, net of issuance costs	20,268,245	—
Repayments of line of credit	(1,386,697)	(299,077)
Repayments of notes payable	(851,845)	(293,104)
Repayments of notes payable – related party	(1,849,888)	(453,388)
Repayments of convertible note payable – related party	—	(3,226,548)
Net cash provided by (used in) financing activities	42,205,404	(2,896,954)

NET CHANGE IN CASH AND RESTRICTED CASH	28,179,799	9,817
CASH AND RESTRICTED CASH, beginning of period	11,712,493	3,970,466
CASH AND RESTRICTED CASH, end of period	$39,892,292	$3,980,283
SUPPLEMENTAL CASH FLOW INFORMATION
Cash paid for income taxes	$—	$—
Cash paid for interest	$2,229,681	$301,593
Non-cash financing and investing activities
Non-cash contribution to equity method investment	$—	$115,210
Issuance of Common Stock and Series D Preferred Stock for the acquisition of TCM	$—	$20,250,013
Exchange of TCM stock options resulting in issuance of stock options in acquisition	$—	$1,883,955
Issuance of preferred stock upon partial conversion of convertible note	$—	$164,427
Acquisition of right-of-use asset in exchange for lease obligations	$45,504,172	$5,820,225
Lease liabilities arising from obtaining right-of-use assets	$43,713,270	$6,078,929

Reconciliation of Net Income (Loss) to Adjusted EBITDA (Unaudited)

The accompanying press release refers to Adjusted EBITDA, a non-GAAP financial measure not calculated or presented in accordance with GAAP. This non-GAAP financial measure is supplemental information and in addition to the financial measures presented in the accompanying release that are calculated and presented in accordance with GAAP.

EBITDA is defined as net income (loss) before interest expense, net; income tax expense (benefit); and depreciation and amortization. Adjusted EBITDA is defined as EBITDA further adjusted to exclude stock-based compensation expense and gains and losses arising from the issuance, remeasurement, extinguishment or disposition of financial instruments, investments and other assets.

Management uses Adjusted EBITDA to evaluate operating performance, establish budgets and forecasts and make operational decisions. The company believes the measure is useful to investors because it excludes items that management does not consider indicative of the underlying operating performance of the business, and because it facilitates comparison of results across periods.

Adjusted EBITDA has limitations as an analytical tool and should not be considered in isolation or as a substitute for net income (loss) or any other measure prepared in accordance with GAAP. Among other limitations, Adjusted EBITDA excludes depreciation and amortization of assets that are central to the company's ability to generate revenue, and excludes interest expense and amortization associated with the company's finance lease obligations, under which the company expects to make payments of approximately $9.3 million during the remainder of 2026 and approximately $20.0 million during 2027. Adjusted EBITDA as defined by the company may not be comparable to similarly titled measures reported by other companies.

	For the Three Months Ended June 30,		For the Six Months Ended June 30,
	2026	2025	2026	2025
Revenue	$6,712,805	$3,085,439	$10,132,723	$4,956,691

Net income (loss)	(22,776,206)	12,118,574	(72,392,917)	10,296,097

Depreciation and amortization (inclusive of ROU amortization)	6,882,154	1,059,900	9,548,999	1,537,948
Interest expense, net	2,045,748	601,260	2,584,024	940,946
Income tax expense (benefit)	(63,154)	285,120	(185,064)	285,120
Stock based compensation	119,939	233,208	241,809	250,552
Change in fair value of warrant liability	—	—	1,585,838	692,103
Change in fair value of digital assets	—	(85,756)	—	(37,984)
Change in fair value of convertible note	(2,380,000)	—	(2,380,000)	—
Change in fair value of additional convertible notes option	(3,850,850)	—	(3,850,850)	—
Gain on sale of equity method investments	—	—	(12,569,661)	—
Gain on disposal of property and equipment	(1,034)	—	(36,298)	—
Loss on issuance of convertible note	19,241,000	—	73,881,850	—
Loss on extinguishment of debt	—	71,094	—	153,834
Gain on remeasurement of investment in TCM	—	(14,549,536)	—	(14,549,536)
Adjusted EBITDA	$(782,403)	$(266,136)	$(3,572,270)	$(430,920)

Adam Brown 8/20/26 Adam Brown 8/20/26

QumulusAI Anchors Metro Atlanta Data Center Site With a Contracted 3.75 MW and a Path to 10.75 MW

The seven-year agreement covers up to 3.75 MW today, with a right of first offer for up to 7 MW of additional contiguous capacity.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced a seven-year colocation agreement for up to 3.75 MW of data center capacity in metropolitan Atlanta, together with a right of first offer on as much as 7 MW of additional contiguous capacity at the same site. The contracted capacity is expected to support up to 2,048 NVIDIA Blackwell B300-class GPUs.

Delivery is conditioned on the provider completing the transition of the premises with the existing occupant, and capacity is expected to be ready for service in Q4 2026.

“Contracted demand is only as good as the power you have to serve it,” said Michael Maniscalco, CEO of QumulusAI. “Acquiring multiyear capacity in our own backyard will give us room to land deployments we are signing now, and a first look at the room to grow next door.”

The agreement addresses the supply side of a demand picture QumulusAI has been building publicly since early June. The company has announced customer agreements valued at more than $246 million in aggregate over that period, including a three-year inference agreement valued at more than $71 million, announced July 28, and a separate GPU-as-a-Service agreement with DRW, announced Aug. 11. In July, it purchased 1,632 NVIDIA Blackwell B300 GPUs to meet that demand. Atlanta is also QumulusAI’s home market; the company established its corporate headquarters in Georgia Tech’s Tech Square in June.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company’s colocation agreement for up to 3.75 MW of data center capacity in metropolitan Atlanta, the anticipated term of the agreement and the final committed capacity to be established under it, the satisfaction of the conditions to delivery — including the existing occupant’s transition out of the premises — the expected timing of phased delivery and readiness for service, the company’s ability to sign and host additional customer deployments using the contracted capacity and room to grow within the contracted capacity, the anticipated number of NVIDIA Blackwell B300-class GPUs the site is expected to support, whether the provider decides to expand contiguous capacity at the site and the timing of any such expansion, the potential combined capacity of as much as 10.75 MW that could become available to the company at the site if that expansion is built and the company’s right of first offer is exercised, and the value that the company may realize from existing customer agreements. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management’s current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company’s dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company’s substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company’s limited operating history and history of net losses; and those described in the “Risk Factors” section of the company’s registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company’s subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 8/18/26 Adam Brown 8/18/26

What QumulusAI’s NVIDIA Cloud Partner Status Means for Customers

QumulusAI is an NVIDIA Cloud Partner. That designation means our AI infrastructure can be built, validated, and operated at the standard of excellence NVIDIA sets for production AI. It's the same standard applied across the entire NVIDIA Partner Network. What varies between partners is how that reference architecture is delivered to customers. For us, that comes down to how quickly capacity comes online, how much of it is dedicated to you, and how closely the hardware maps to your unique use case.

What NCP Status Guarantees

NCP status sets the baseline for how the compute is built and run:

Built and validated against NVIDIA reference architectures. For reference design compliant NVIDIA Blackwell clusters (HGX B200 or HGX B300) the NVLink domain and inter-node fabric are validated against NVIDIA's architecture.
Tested before it enters service. Field-level diagnostics and burn-in across fabric, memory, thermals, and GPU health, measured against a known-good baseline.
Capacity aligned with NVIDIA’s platform roadmap. Next-generation systems available closer to launch schedules, rather than waiting for them to reach the resale market.
Operated continuously. GPU health telemetry runs while workloads are live, firmware is managed across its lifecycle, and nodes are wiped clean between tenants, so no prior tenant's data, keys, or credentials carry forward.

This baseline is consistent across NVIDIA Cloud Partners. The rest is how we build on that foundation to bring even greater value to our clients.

Capacity on Your Timeline

We own and operate our infrastructure across a distributed network of vetted colocation sites with power already in place. Because the sites and power are already secured, capacity comes online faster than a ground-up data center build. It can scale up for a training run or a shift in inference demand with our Capacity Planning as a Service.

Dedicated Infrastructure

Compute is dedicated and single-tenant, with regional placement and private-cloud control. For inference platforms managing their own margins, and for enterprise teams with data-residency or control requirements, dedicated infrastructure is a practical requirement rather than a preference.

A Fleet Matched to the Workload

We run current-generation NVIDIA hardware and match it to the job:

NVIDIA B300 / B200 GPUs (Blackwell) — large-scale training and high-throughput, low-latency inference.
NVIDIA H200 / H100 GPUs, SXM and NVL (Hopper) — production training and serving; NVIDIA H200 GPUs for memory-bandwidth- and context-heavy inference.
NVIDIA RTX PRO 6000 — cost-efficient production inference and mixed graphics/compute.

Reference design-complaint NVIDIA NVL8 clusters are available for the largest workloads. Smaller or more specialized jobs run on the configuration that fits them.

Working with QumulusAI

NCP status establishes that the compute meets NVIDIA's standard. Our model determines how fast you can get it, how much of it is yours, and how well it fits the work.

To discuss specifics against your workload, contact us.

FAQ

An NVIDIA Cloud Partner is a cloud provider within the NVIDIA Partner Network (NPN) that delivers hosted infrastructure built on the NVIDIA accelerated computing platform and purpose-built for AI workloads at production scale. Unlike a general-purpose cloud with GPUs added on, an NCP's infrastructure is engineered specifically for training, fine-tuning, and inference.
The NVIDIA Partner Network is NVIDIA's global partner program, spanning resellers, solution providers, service-delivery partners, and cloud providers. NVIDIA Cloud Partner is the cloud-provider designation within the NPN, focused on hosted, high-performance AI infrastructure.
DescriIt means a provider's infrastructure is built and validated against NVIDIA's reference architectures and engineered for production AI. That covers how systems are configured — GPUs, the NVLink domain, and the inter-node networking fabric — how each node is validated before it enters service, and how the environment is operated for reliability at scale. It is a consistent baseline standard applied across partners.ption text goes here
A general GPU cloud rents access to GPUs. An NVIDIA Cloud Partner goes further: its infrastructure is built and validated against NVIDIA's reference architectures for AI specifically, so the whole system — not just the GPU, but the interconnect, fabric, and operational tooling — is designed to deliver consistent, predictable performance at production scale.
NVIDIA Cloud Partners support the full range of production AI: large-scale model training, fine-tuning, reinforcement learning, and high-throughput, low-latency inference. We serve both enterprise AI teams and inference/AI-platform providers running at scale.
All NVIDIA Cloud Partners meet the same baseline for how compute is built and operated. QumulusAI's difference is in delivery: we own and operate our infrastructure across a distributed network of vetted colocation sites with power already in place, which supports dedicated capacity, regional control, and shorter lead times than a ground-up build.
Faster than a ground-up build. Because we deploy into vetted colocation sites where power is already secured, adding capacity takes new data center construction out of the timeline.
Yes. QumulusAI provides dedicated infrastructure with private-cloud control, regional placement, and clean isolation between tenants — no prior tenant's data, keys, or credentials carry forward.
NVIDIA Blackwell (B300, B200), NVIDIA Hopper (H200, H100 in SXM and NVL), and NVIDIA RTX PRO 6000, matched to the workload. We add the latest NVIDIA platforms as they become available, including Vera Rubin.

Adam Brown 8/17/26 Adam Brown 8/17/26

QumulusAI to Report Second Quarter 2026 Financial Results on August 25, 2026

The company will host a conference call and webcast at 5 p.m. EDT the same day to discuss results for the quarter ended June 30, 2026.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced that it will report financial results for the second quarter ended June 30, 2026 after market close on Tuesday, August 25, 2026, and will host a conference call and webcast at 5 p.m. EDT the same day to discuss the results.

The call will be the company’s first quarterly results call as a public company, following the start of trading on the Nasdaq Global Market under the ticker symbol “QMLS.”

Conference Call and Webcast Details

Date: August 25, 2026
Time: 5 p.m. EDT
Webcast:https://events.q4inc.com/attendee/772707480
Domestic dial-in: 833-461-5787
International dial-in: +1 585-542-9983
Meeting ID: 772 707 480

Participants are encouraged to join 10 minutes before the call starts. A live webcast and an archived replay will be available in the Investor Relations section of the company’s website at investors.qumulusai.com.

About QumulusAI

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 8/11/26 Adam Brown 8/11/26

QumulusAI Signs GPU-as-a-Service Agreement With DRW for NVIDIA Blackwell B300 Capacity, Annually Renewable Up to Four Years

DRW, a diversified global trading firm, will use a dedicated NVIDIA Blackwell B300 cluster to expand the scale and speed of its compute resources under an annually renewable agreement.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced a GPU-as-a-Service agreement with DRW, a diversified global trading firm with expertise in commodities, derivatives and emerging asset classes, to supply a dedicated NVIDIA Blackwell B300 cluster.

DRW has been an early participant in the development of compute as an institutional market. DRW incubated Compute Exchange, a marketplace connecting enterprises seeking GPU capacity with providers of compute infrastructure, and Silicon Data, which has developed pricing benchmarks, indices, and futures contracts for the GPU compute market.

“Quantitative trading firms are among the most demanding compute customers there are, and they are expanding what they invest in AI infrastructure,” said Michael Maniscalco, CEO of QumulusAI. “GPU-as-a-Service will give DRW fast, dedicated access to current-generation Blackwell compute exactly when its teams need it. We are glad to welcome DRW as a customer.”

Capacity will be served from QumulusAI’s U.S. data center footprint. The agreement carries an initial one-year term and three one-year renewal options, and a contractual framework that can run up to four years in total.

“We’ve always invested in the research and technology that allow us to test, challenge and refine how we understand markets,” said Rich Norman, Chief Information Officer of DRW. “This capacity allows us to run more complex research at greater scale, iterate faster and shorten the distance between a hypothesis and what the data tells us. That ability to keep learning is fundamental to finding new edge.”

The agreement follows QumulusAI’s August 7 announcement of an agreement with an agentic hedge fund and brings a second financial-markets customer into a book of business led to date by AI inference platforms. QumulusAI has announced customer agreements valued at more than $246 million in aggregate since early June.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company’s GPU-as-a-Service agreement with DRW for NVIDIA Blackwell B300 capacity, the potential term of the agreement and whether its annual renewal options are exercised, DRW’s ability to use the capacity to expand the scale and speed of its compute resources and research, the company’s ability to provide fast, dedicated access to current-generation Blackwell compute exactly when DRW’s teams need it, and the company’s ability to serve the agreement from its existing U.S. data center footprint. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management’s current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company’s dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company’s substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company’s limited operating history and history of net losses; and those described in the “Risk Factors” section of the company’s registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company’s subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 8/7/26 Adam Brown 8/7/26

QumulusAI Signs Agreement With Agentic Hedge Fund To Provide NVIDIA Blackwell GPU Capacity

The agreement marks the company’s first deployment of a new monetization strategy for its reserve NVIDIA Blackwell GPU capacity, pairing market-rate compute revenue with a share of the customer’s trading profits and no exposure to trading losses.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced an agreement with an agentic hedge fund. QumulusAI negotiated market-rate compute pricing for the fund’s usage along with a share of its quarterly trading profits, with no exposure to trading losses.

The customer operates a fully agentic hedge fund whose specialized AI agents discover, test, validate and deploy trading strategies with live capital, around the clock, on QumulusAI’s self-hosted, sovereign compute environment utilizing NVIDIA Blackwell GPU capacity to support the fund’s around-the-clock trading operations.

This agreement represents the first deployment of a new monetization strategy for those assets, combining market-rate compute revenue with a share of trading profits while providing no exposure to trading losses. The structure differs from the fixed-value, take-or-pay agreements QumulusAI has announced in recent weeks: rather than a set contract value, revenue under this agreement will vary with both the compute the fund consumes and the trading performance it generates within the profit-sharing threshold. While there can be no assurance that the strategy will improve profitability, management believes it has the potential to increase the economic value QumulusAI generates from its reserve capacity over time.

“Financial markets move in milliseconds, and those trading them need infrastructure that can keep pace,” said Michael Maniscalco, CEO of QumulusAI. “A fully agentic hedge fund that is discovering, testing, validating and acting on strategies around the clock needs self-hosted, sovereign compute built for that pace.”

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the agreement and its anticipated terms — including the compute pricing and the profit-sharing arrangement negotiated with the customer — the company’s ability to serve the agreement’s NVIDIA Blackwell GPU capacity requirements from its self-hosted, sovereign compute environment, management’s belief regarding the potential of its new reserve-capacity monetization strategy to increase the economic value generated from that capacity over time, and the customer’s continued use of that environment for its trading operations. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 8/6/26 Adam Brown 8/6/26

QumulusAI to Participate in Needham Virtual AI Infrastructure 1x1 Conference

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced that CEO Mike Maniscalco will participate in the Needham Virtual AI Infrastructure 1x1 Conference on Wednesday, Aug. 12, 2026.

Management will host virtual one-on-one meetings with institutional investors throughout the day to discuss the company's hyper-distributed compute deployments and recent milestones in scaling cost-efficient infrastructure for AI inference platforms. Meetings will be conducted in 40-minute intervals.

"A lot has happened in a short window. We have announced over $240 million in compute contracts since June, and we are already on the ground delivering 6 customer deployments — several of them close to done, with others finishing this quarter," said Mike Maniscalco, CEO of QumulusAI. "That is what I want investors to hear next week. Compute does not have to take years to stand up, and it does not have to come from a hyperscaler. We are proving there is a faster way."

Institutional investors wishing to schedule a one-on-one meeting with QumulusAI management should reach out to their Needham representative.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the value of the more than $240 million in customer contracts announced since June 2026, the status and expected timing of completion of the six customer deployments underway under those contracts, and the company’s ability to make compute available faster. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 7/31/26 Adam Brown 7/31/26

QumulusAI Rings the Nasdaq Opening Bell

Photos and video from QMLS's Times Square celebration, Credit: 2026 Nasdaq, Inc. / Vanja Savic

QumulusAI (Nasdaq: QMLS) rang the Nasdaq Stock Market opening bell at the Nasdaq MarketSite in Times Square on Wednesday, July 29, 2026, marking the company's formal welcome to the exchange two weeks after its July 16 direct listing. CEO Michael Maniscalco rang the bell alongside employees, board members, and partners who traveled to New York for the occasion.

David Wicks, Nasdaq's Vice President of Listing, opened the ceremony by welcoming QumulusAI into the exchange's roster of companies. "Today we proudly celebrate QumulusAI's listing on Nasdaq, and we couldn't be more proud to welcome you all to our family of the world's leading companies," said Wicks. "Nasdaq is incredibly proud to be your partner, and we look forward to supporting your continued growth and innovation as a Nasdaq-listed company."

Michael Maniscalco, CEO of QumulusAI, used the moment to thank the people who built the company from its early days to its most recent. "I want to thank our employees who got us to this point; our shareholders who believed in us early; our partners who helped us build the ecosystem and bring the infrastructure to life for our customers; our customers who are trusting us to get their infrastructure online — their critical infrastructure online — and keep it running," said Maniscalco.

Credit: Nasdaq, Inc.

He also made the case for why QumulusAI exists in the first place. "The solutions that we're providing are bringing the AI applications and models to life. We build the infrastructure that powers the AI. And when I look at the space, and I look at all the incredible things happening around it, I can't think of a better space to be in," Maniscalco said.

The weeks around the listing carried their own momentum: QumulusAI announced a $71.9 million, three-year agreement with an AI inference platform provider on July 28, a $32 million Blackwell B300 agreement on July 23, and an $18 million take-or-pay agreement with a GPU marketplace partner on July 22.

Adam Brown 7/28/26 Adam Brown 7/28/26

QumulusAI Signs $71.9 Million, Three-Year Agreement With AI Inference Platform Provider

The customer, a platform that helps companies deploy and serve AI models in production, will draw on NVIDIA Blackwell B300 and B200 supply from QumulusAI's AI cloud GPU capacity under a three-year agreement with renewal options.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced a signed three-year agreement, valued at more than $71 million, to supply NVIDIA Blackwell B300 and B200 capacity to an AI inference platform provider. The agreement includes renewal options and ranks among the company's largest customer commitments announced to date.

The customer's platform helps companies deploy large language models, vision models, speech models and other AI applications with low latency and high reliability and runs models for customers specializing in LLMs, image and video generation. The dedicated Blackwell capacity will give the platform high-performance compute to serve production inference workloads as demand from its customers scales.

Capacity under the agreement will be served from QumulusAI's U.S. data center footprint and is expected to be ready for customer use in the third quarter of 2026. The company's demand-led deployment model places capacity into available pockets of power across a distributed network of colocation and owned facilities, enabling it to bring GPU capacity online in months, not years. The agreement adds more than $71 million in contracted, multiyear commitments to QumulusAI's book of business.

“Inference is where AI meets the real world, and the platforms serving it can't afford to wait on capacity,” said Michael Maniscalco, CEO of QumulusAI. “Our customer runs production workloads for companies that need the right combination of flexibility, access, cost, trust and speed in their infrastructure. A three-year commitment of this size reflects what our model is built to do: procure and deploy Blackwell capacity where demand already exists, and do it fast.”

The agreement extends a series of recent demand announcements. On July 23, QumulusAI announced a $32 million, two-year NVIDIA Blackwell B300 agreement with an AI inference platform provider focused on generative AI applications. On July 22, the company announced a two-year, take-or-pay Blackwell B300 agreement valued at more than $18 million with a GPU cloud marketplace serving AI teams in more than 100 regions worldwide. On June 11, it announced $124.4 million in three-year inference agreements across two customers, and on May 28, it announced two NVIDIA H200 cluster deployments with marketplace partner Shadeform. Together with today's announcement, the company has announced more than $246 million in customer agreements since early June.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company's agreement, the anticipated value and term of the agreement, for Blackwell B300 and B200 capacity to serve production inference workloads, the company's ability to serve the agreement from its existing data center footprint, the expected timing of deployment of NVIDIA Blackwell capacity, and the pace at which the company can deploy capacity and bring GPU compute online. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 7/27/26 Adam Brown 7/27/26

QumulusAI Set to Ring the Nasdaq Opening Bell

Company will ring the opening bell at the Nasdaq MarketSite in New York's Times Square on Wednesday, July 29.

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced it will ring the opening bell at the Nasdaq MarketSite in New York's Times Square on Wednesday, July 29. The live broadcast will begin at 9:15 a.m. EDT on nasdaq.com/marketsite/bell-ringing-ceremony.

"Ringing the Nasdaq bell is a milestone moment in a company's life, and it is a privilege to mark ours," said Michael Maniscalco, CEO of QumulusAI. "We are grateful to the employees, board members, investors, partners and customers who helped us reach this point, and it means a great deal to celebrate it together on Nasdaq's stage."

About QumulusAI

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 7/23/26 Adam Brown 7/23/26

QumulusAI Signs $32 Million, Two-Year NVIDIA Blackwell B300 Agreement With AI Inference Platform Provider

The customer, an inference platform serving production generative AI applications, will run image, video and other generative workloads on dedicated Blackwell B300 clusters from QumulusAI's hyperdistributed AI cloud

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced a signed two-year agreement, valued at more than $32 million, to supply NVIDIA Blackwell B300 capacity to an AI inference platform provider focused on generative AI applications. The agreement includes renewal options, with capacity expected to come online in the fall of 2026.

The customer's platform delivers fast, scalable infrastructure for generative AI applications, serving developers and enterprises that run image, video and other generative models where speed and reliability determine the user experience. Under the agreement, QumulusAI will provide dedicated GPU clusters that give the platform committed, high-performance capacity as demand from its customers scales.

Serving generative media workloads at scale typically calls for sustained, dedicated compute rather than opportunistic spot capacity — a dynamic reflected in multiyear commitments such as this agreement. Capacity will be served from QumulusAI's U.S. data center footprint. The company's demand-led deployment model places capacity into available pockets of power across a distributed network of colocation and owned facilities, enabling it to bring GPU capacity online in months, not years.

The agreement adds a two-year commitment of more than $32 million to QumulusAI's book of business.

“Generative media workloads put real pressure on inference infrastructure — images and video are compute-intensive to serve, and the user experience depends on speed,” said Mike Maniscalco, CEO of QumulusAI. “This agreement reflects a pattern we're seeing in our own business: inference customers want dedicated, committed capacity they can count on, and our model is built to put that capacity to work quickly.”

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company's agreement, the anticipated value and term of the agreement, the expected timing of deployment of NVIDIA Blackwell B300 capacity, the company’s ability to provide dedicated GPU clusters that give the platform committed, high-performance capacity, the company's ability to serve the agreement from its existing data center footprint, and the pace at which the company can deploy capacity and bring GPU compute online. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Source: QumulusAI

Adam Brown 7/22/26 Adam Brown 7/22/26

QumulusAI Signs $18 Million, Two-Year Take-or-Pay NVIDIA Blackwell B300 Agreement With Marketplace Partner

The customer, a GPU cloud provider serving AI teams in more than 100 regions worldwide, will source NVIDIA Blackwell B300 nodes from QumulusAI under a two-year, take-or-pay agreement

ATLANTA--(BUSINESS WIRE)-- QumulusAI, a neocloud infrastructure provider purpose-built for the AI computing era, today announced a two-year, take-or-pay agreement to supply NVIDIA Blackwell B300 nodes to a GPU cloud marketplace used by AI teams across more than 100 regions worldwide. The contract, valued at more than $18 million, makes QumulusAI a core Blackwell supplier to the customer.

Capacity under the agreement will be served from QumulusAI's active U.S. data center footprint, with initial deployments expected to come online this summer. The company's demand-led deployment model places capacity into available pockets of power across a distributed network of colocation sites, enabling it to bring GPU capacity online in months, not years.

“Our customers are building the AI economy, and they choose their suppliers carefully,” said Mike Maniscalco, CEO of QumulusAI. “A two-year commitment for our Blackwell capacity says it plainly. The demand is real, and we are here to deliver.”

The agreement adds a two-year, take-or-pay commitment of more than $18 million to the company's book of business.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company's agreement, the anticipated value and term of the agreement and the revenue expected under it, the expected timing of initial deployments of NVIDIA Blackwell B300 capacity, the company's ability to serve the agreement from its existing data center footprint, the pace at which the company can deploy capacity and bring GPU compute online, and future performance. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “can,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the “Risk Factors” section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 7/20/26 Adam Brown 7/20/26

QumulusAI Purchases 1,632 NVIDIA Blackwell B300 GPUs Amid Strong Customer Demand

Dual-sourced order of 204 NVIDIA HGX B300 systems expands the company's hyperdistributed AI cloud as customer demand accelerates

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced the purchase of 1,632 NVIDIA Blackwell B300 GPUs to meet accelerating customer demand for high-performance AI compute. The GPUs will be delivered across 204 NVIDIA HGX B300 systems and represent one of the company's largest single capacity expansions to date. The purchase was funded primarily through financing arrangements with Technology Finance Corporation and USD.ai.

The order lands as demand for AI infrastructure continues to outpace supply. Against that backdrop, QumulusAI has been converting signed, multiyear customer demand into deployed compute, including more than $124 million in three-year inference agreements announced earlier this year. The B300 purchase directly supports that contracted and near-term pipeline.

"Meeting AI demand today is a test of access, speed and flexibility, and you can't win on one without the other two," said Mike Maniscalco, CEO of QumulusAI. "Our model gives us the flexibility to source across multiple OEM partners such as Supermicro and Lenovo and to deploy across a national network of colocation and owned facilities. When demand moves, we move with it. That's how 1,632 B300 GPUs go from purchase order to production at hyperspeed."

Alongside the B300 systems, which are built for large-scale AI training and high-throughput inference, the order includes an additional 192 NVIDIA RTX PRO 6000 Blackwell GPUs. Together, the two architectures give QumulusAI customers workload-matched options spanning training, inference and visualization, consistent with the company's F.A.C.T.S. approach of mapping the right compute to each AI workload rather than forcing every job onto a single platform.

The new capacity extends a period of rapid growth for QumulusAI, whose deployed GPU fleet has expanded by more than 450% from June 2025 to June 2026.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company’s ability to source across multiple OEM partners like Supermicro and Lenovo and to deploy across a national network of colocation and owned facilities. Words such as "anticipate," "believe," "estimate," "expect," "guidance," "intend," “can,” "may," "on track," "plan," "project," "target," "will" and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the "Risk Factors" section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Source: QumulusAI

Adam Brown 7/17/26 Adam Brown 7/17/26

QumulusAI Joins the NVIDIA Partner Network as an NVIDIA Cloud Partner

ATLANTA--(BUSINESS WIRE)-- QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced it has been approved as an NVIDIA Cloud Partner (NCP) within the NVIDIA Partner Network (NPN), reinforcing its ability to bring high-performance compute online quickly to meet growing customer demand.

As an NVIDIA Cloud Partner, QumulusAI can work with AI-native companies, enterprises, and machine learning teams to deploy NVIDIA AI infrastructure for modern AI workloads spanning model training, fine tuning, reinforced learning, and production-scale inference.

Enterprises and AI teams need compute across the full spectrum of AI workloads — from large-scale model training to scaling inference and autonomous AI agents — and they need that capacity delivered faster than traditional infrastructure developers can keep pace. QumulusAI rapidly expands available capacity by deploying infrastructure across a distributed network of vetted co-location data center sites with access to power, enabling the company to bring high-performance compute online in accelerated timeframes rather than waiting on large, multi-year data center developments. This lets customers access the capacity they need now, and scale it as their AI ambitions grow.

That speed-to-capacity is anchored in QumulusAI's F.A.C.T.S. framework — Flexibility, Access, Cost, Trust, and Speed — the foundation of a business built to break down the barriers customers face in building and adopting AI.

"Right now, our customers need capacity, and they need it fast. The neoclouds that win are the ones that can unlock capacity and put it to work for clients as demand accelerates," said Michael Maniscalco, CEO of QumulusAI. "Becoming an NVIDIA Cloud Partner affirms what our customers already experience — enterprise-grade NVIDIA compute, delivered at hyperspeed. It's about giving customers what they need today, while building toward a larger inference-future we believe is coming."

About QumulusAI

Follow QumulusAI on LinkedIn and X @QumulusAI.

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the impact of being an NVIDIA Cloud Partner and building toward a larger inference future. Words such as "anticipate," "believe," "estimate," "expect," "guidance," "intend," “can,” "may," "on track," "plan," "project," "target," "will" and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the "Risk Factors" section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Investor Contactinvestors@qumulusai.com

Media Contactmedia@qumulusai.com

Adam Brown 7/16/26 Adam Brown 7/16/26

A Letter to Our Investors and Partners

To our investors and partners,

We started QumulusAI on a simple conviction: as AI moves from research to production, the winners won’t be whoever buys the most GPUs. They’ll be whoever can bring AI infrastructure online with discipline, transparency, and speed. That conviction drives everything we do, and the momentum behind it has never been stronger.

It comes down to a distinction that sits at the heart of our company. AI’s biggest barrier isn’t intelligence; it’s the compute that enables it. We don’t build the models. We build the infrastructure to power them. I’m writing to show you what we’ve built, how fast we’re building it, and why we’re positioned to reshape how organizations access AI compute.

The Problem We Exist to Solve

Demand for AI compute is outpacing the infrastructure to support it, and we are still early. Much of today’s compute is cornered by the largest AI companies and hyperscalers, whose pricing runs artificially high on internal supply constraints. Organizations scaling their AI use cases hit the same walls: fragmented procurement, opaque pricing, and environments that limit where they can run workloads. We see five structural barriers: flexibility, access, cost, trust, and speed. We organize everything around solving them. That’s our F.A.C.T.S. framework, not marketing jargon but the lens we apply to every site, every deployment, and every customer engagement.

What We’re Building

What we’re building comes down to a straightforward goal: bring as much AI compute online as fast as possible. That’s what the market needs right now. Customers don’t want capacity eventually; they want it yesterday. Supporting this surging demand is largely a question of speed, and speed is where a distributed approach wins. A single mega-campus can take years to plan, power, and energize. We bring AI compute online in months, not years. We use existing capacity to meet demand now, while simultaneously building the pipeline to power what’s next, running compute across a growing network of smaller sites and standing up fully operational GPU-as-a-Service environments at hyperspeed. And where and when it makes sense, we will leverage full-stack ownership across power, GPU deployment, and orchestration to control that timeline ourselves rather than wait on others to hit it.

Our model is already running and accelerating. We have over 3,000 GPUs, deployed and being delivered, generating revenue across our distributed data center network, up from roughly 560 a year ago. Demand is showing up in signed, multi-year commitments: a two-year agreement with Shadeform for two NVIDIA H200 clusters totaling 680 GPUs, and more than $124 million in three-year contracts covering 1,280 NVIDIA Blackwell GPUs across two leading AI inference platforms. That is hyperspeed in practice: capacity identified, deployed, and converted into revenue-generating infrastructure.

We are scaling our next phase of HPC capacity with asset-aligned commitments, including a subsequent $26.0 million equipment lease facility with TFC, $90 million in convertible notes from ATW Partners, and, subject to USD.ai underwriting standards and specific equipment requirements, up to $500 million in non-recourse equipment financing under the USD.AI protocol. Our footprint spans colocation sites across the U.S., including the Atlanta area, Kansas City, Denver, and Philadelphia, with more than 150 MW of available capacity and accelerating deployment cycles. And it all runs on NVIDIA technology. The point isn’t simply that we’re growing. It’s that we’ve built a repeatable way to bring compute online at speed, at a moment when execution discipline and capital efficiency matter as much as scale. We stay nimble, finding small, often-overlooked pockets of datacenter and power capacity and moving on them to deliver what the market needs now. Rather than build our own generation, we follow a deliberate power strategy, and we will tap behind-the-meter and independent power sources where and when it makes sense, which lets us move fast today while unlocking longer-term capacity for scale. Our aim is durable, well-capitalized growth, not growth for its own sake.

Why We’re Confident

The momentum since my first day as CEO in September reflects that confidence. We’ve strengthened our board and leadership as we move toward the public markets, adding directors whose experience spans public-company governance, capital markets, and regulatory strategy: Dr. Homaira Akbari, a seasoned public-company director; David Rench, former CFO of Applied Digital; Barry Schwartz, who brings deep regulatory expertise; and Mike Mulica, a veteran technology CEO and venture capitalist with three decades of experience in scaling technology, communications, internet, and cloud platforms. We grew our GPU capacity by 5.5x during our 2025 expansion phase, much of it under multi-year customer contracts, with future growth potential captured in our recently released guidance. And speed remains our edge: smaller, distributed sites bring compute online faster than monolithic campuses, and as workloads shift from training toward inference, that footprint becomes a lasting structural advantage across a wide range of future AI workloads.

We’ve proven QumulusAI can identify capacity, procure and deploy GPUs, stand up operational environments quickly, and convert infrastructure into customer-ready AI compute. The AI revolution will be built on infrastructure, and we intend to build the layer that helps more organizations break AI’s biggest barriers, accessing the compute they need faster, more transparently, and with greater flexibility.

None of this would exist without our people. Our early founders took a bold idea and willed it into a company, and every employee since has carried it forward through persistence, hard work, and vision. What we’ve built is a credit to them, and it’s the foundation we’ll keep building on.

To our legacy investors: thank you for believing early. The company you backed has grown from an ambitious idea into an operating platform with deployed GPUs, revenue-generating infrastructure, and a clear strategy to capitalize on one of the biggest infrastructure buildouts in history. To new investors and partners: we’d welcome the chance to show you what we’re building, a differentiated model, a large market opportunity, and a team built to execute.

We are not hyperscale as you know it. We are building for hyperspeed.

Thank you for your continued trust and partnership. I look forward to sharing more as we execute.

Sincerely,
Mike Maniscalco
Chief Executive Officer, QumulusAI

Adam Brown 7/16/26 Adam Brown 7/16/26

QumulusAI Begins Trading on the Nasdaq Global Market Under Ticker Symbol “QMLS”

Inference-first AI cloud platform enters public markets to accelerate distributed compute deployment

ATLANTA--(BUSINESS WIRE)--QumulusAI (Nasdaq: QMLS), a neocloud infrastructure provider purpose-built for the AI computing era, today announced that its common shares have begun trading on the Nasdaq Global Market under the ticker symbol QMLS.

“QumulusAI was built for how AI is actually deployed — inference-first, distributed, and close to demand,” said Mike Maniscalco, CEO of QumulusAI. “As an active, revenue-generating neocloud, we bring compute online in months, not years. Becoming a public company gives us the platform, and the capital efficiency, to scale that model as enterprise demand for AI infrastructure compounds.”

QumulusAI addresses the critical shortage in AI compute through a highly flexible, distributed infrastructure strategy. The company rapidly deploys and monetizes GPU infrastructure for AI inference, training, and agentic workloads, while simultaneously advancing vertically integrated approaches to support massive-scale enterprise workloads.

Copies of the final prospectus relating to the registration may be obtained by visiting the SEC's website at www.sec.gov.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the company's ability to scale its model. Words such as "anticipate," "believe," "estimate," "expect," "guidance," "intend," "may," "on track," "plan," "project," "target," "will" and similar expressions are intended to identify forward-looking statements. These statements are based on management's current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company's dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company's substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company's limited operating history and history of net losses; and those described in the "Risk Factors" section of the company's registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company's subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Contacts

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 7/15/26 Adam Brown 7/15/26

QumulusAI to Begin Trading Tomorrow on the Nasdaq Global Market Under Ticker Symbol “QMLS”

U.S. Securities and Exchange Commission declares Form S-1 registration statement effective; trading expected to begin Thursday, July 16, 2026, through a direct listing

ATLANTA--(BUSINESS WIRE)--QumulusAI, a neocloud infrastructure provider purpose-built for the AI computing era, today announced that it expects its common stock to begin trading tomorrow, Thursday, July 16, 2026, on the Nasdaq Global Market under the ticker symbol “QMLS.” The direct listing follows the U.S. Securities and Exchange Commission declaring the company's registration statement on Form S-1 effective on July 14, 2026.

"Listing on the Nasdaq marks a transformative milestone for QumulusAI as we enter our next phase of growth," said Michael Maniscalco, CEO of QumulusAI. "We believe AI demand continues to outpace infrastructure supply, and we designed our hyperspeed and capital-efficient model to close that gap. This direct listing provides us with the platform to scale available infrastructure, engage with a broader investor base and continue delivering accelerated compute to the enterprises building the future of AI."

Chardan Capital Markets LLC is acting as the company’s financial advisor in connection with the direct listing.

New investor materials, including a comprehensive investor presentation outlining the company's business model, are available on QumulusAI's recently launched investor relations website at investors.qumulusai.com.

About QumulusAI

Forward-Looking Statements

This press release contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding the expected commencement of trading of the company’s common stock on the Nasdaq Global Market on July 16, 2026, the company’s growth, AI demand, and the company’s ability to scale available infrastructure, engage with a broader investor base and continue delivering accelerated compute to the enterprises building the future of AI. Words such as “anticipate,” “believe,” “estimate,” “expect,” “guidance,” “intend,” “may,” “on track,” “plan,” “project,” “target,” “will” and similar expressions are intended to identify forward-looking statements. These statements are based on management’s current expectations and assumptions as of the date of this release and are subject to risks and uncertainties that could cause actual results to differ materially, including, among others, the company’s dependence on a limited number of large customers; the availability and cost of power, network connectivity and specialized hardware such as graphics processing units; the company’s substantial capital requirements and access to financing; competition and rapid technological change in the high-performance computing and AI markets; the company’s limited operating history and history of net losses; and those described in the “Risk Factors” section of the company’s registration statement on Form S-1, as amended (File No. 333-292514), filed with the U.S. Securities and Exchange Commission (SEC), as such factors may be updated in the company’s subsequent filings with the SEC. QumulusAI undertakes no obligation to update or revise any forward-looking statement, whether as a result of new information, future developments or otherwise, except as required by applicable law.

Contacts

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Adam Brown 7/14/26 Adam Brown 7/14/26

QumulusAI Issues Fiscal Year 2026 Guidance: $300 Million in Forward ARR, Backed by 18 MW of Capacity

QumulusAI expects approximately 30x ARR growth in fiscal 2026 over fiscal 2025, with line of sight to 2.5 gigawatts in 2027

ATLANTA--(BUSINESS WIRE)--QumulusAI, a neocloud infrastructure provider purpose-built for the artificial intelligence (AI) computing era, today issued fiscal year 2026 guidance in an investor outlook update covering two operational metrics: forward annualized recurring revenue (ARR) and capacity. For fiscal year 2026, the company expects forward ARR of $300 million, approximately 30x growth in fiscal 2026 over fiscal 2025, backed by 18 megawatts (MW) of year-end fiscal 2026 data center capacity with a line of sight to 2.5 gigawatts (GW) by year-end fiscal 2027. A supplemental guidance presentation is available on the company’s newly launched investor relations website at investors.qumulusai.com.

The forward ARR guidance reflects breakout growth. Forward ARR of $300 million is expected as of December 31, 2026, and comprises executed contract revenue to date, expected renewals, deposit-backed compute capacity reservations and projected contract signings, in each case reflecting ARR expected upon activation of the associated compute capacity.

That revenue is backed by continued delivery of compute capacity fueled by customer demand. QumulusAI expects 18 MW of data center capacity as of December 31, 2026, comprising 8 MW of active high-performance computing (HPC) power and 10 MW of HPC power in development, under executed lease or colocation agreements that are operational or undergoing build-out, power activation, or customer deployment. Growth beyond 2026 is supported by line of sight to 2.5 GW of capacity by the end of fiscal year 2027. QumulusAI’s national footprint currently includes active sites in Marietta, Georgia; Kansas City, Missouri; Denver, Colorado; and two sites in Philadelphia, with an Oklahoma site and additional colocation partners in development. Through the QAI Moon joint venture and its agreement with Connected Nation Internet Exchange Points, the company sees a path to 125 network-edge sites across its distributed AI-XP footprint.

“AI’s biggest barrier isn’t intelligence; it’s the infrastructure to power it. We don’t build the models — we build the infrastructure that powers them,” said Michael Maniscalco, CEO of QumulusAI. “Our focus in 2026 is straightforward: bring GPU capacity online in months, not years, and convert it into durable recurring revenue.”

QumulusAI houses the latest NVIDIA GPUs, including Blackwell (B300/B200), Hopper (H100/H200) and RTX PRO 6000, in colocation facilities and data centers of less than 50 MW capacity, on a quarterly deployment cadence designed to accelerate time to value for customers.

“We designed this guidance framework to be simple to track,” said Scott Krosnowski, CFO of QumulusAI. “Forward ARR and capacity capture both sides of our model: the demand we activate and the infrastructure we deliver it on. We are issuing this as annual guidance, and we expect to update our outlook for 2027 at the end of the year.”

Fiscal Year 2026 Guidance

For fiscal year 2026, QumulusAI is providing the following guidance:

  
        Metric
        FY2026 Guidance
      
        Forward Annualized Recurring Revenue (ARR)¹
        $300M (~30x growth in fiscal 2026 over fiscal 2025)
      
        Capacity²
        18 MW (line of sight to 2.5 GW by year-end fiscal 2027)

¹ Forward ARR expected as of December 31, 2026, comprising executed contract revenue to date, expected renewals, deposit-backed compute capacity reservations, and projected contract signings, in each case reflecting ARR expected upon activation of the associated compute capacity. ² Data center capacity under executed lease or colocation agreements, operational or undergoing build-out, power activation, or customer deployment, in each case available for, or expected to become available for, compute deployments as of December 31, 2026.

New Investor Relations Website

QumulusAI has launched its investor relations website at investors.qumulusai.com. The site hosts the supplemental guidance presentation along with a more comprehensive investor presentation outlining the company’s business model, as well as press releases, SEC filings, events and corporate governance information.

About QumulusAI

Forward-Looking Statements

Key Operating Metrics

This press release includes certain operating metrics, including forward ARR and capacity, that are presented as supplemental measures of the company’s performance. Forward ARR and capacity are measures that management uses to evaluate the company’s growth and expansion, and believes they are useful to investors in assessing future operations. ARR is not necessarily indicative of future revenue under U.S. Generally Accepted Accounting Principles (GAAP), and investors should not consider ARR as a substitute for revenue recognized under GAAP or other related measures. These measures may not be comparable to similarly titled measures used by other companies.

Contacts

Investor Contact
investors@qumulusai.com

Media Contact
media@qumulusai.com

Second Quarter 2026 Financial Highlights

Second Quarter Operational Highlights

Condensed Consolidated Statements of Operations (Unaudited)

Condensed Consolidated Balance Sheets

Condensed Consolidated Statements of Cash Flows (Unaudited)

Reconciliation of Net Income (Loss) to Adjusted EBITDA (Unaudited)

What NCP Status Guarantees

Capacity on Your Timeline

Dedicated Infrastructure

A Fleet Matched to the Workload

Working with QumulusAI

FAQ

What is an NVIDIA Cloud Partner (NCP)?

What is the NVIDIA Partner Network (NPN)?

What does NVIDIA Cloud Partner status require?

How is an NVIDIA Cloud Partner different from a regular GPU cloud?

What workloads run on an NVIDIA Cloud Partner?

How is QumulusAI different from other NVIDIA Cloud Partners?

How fast can QumulusAI bring capacity online?

Can I get dedicated, single-tenant capacity?

Which NVIDIA GPUs does QumulusAI run?

Conference Call and Webcast Details

The Problem We Exist to Solve

What We’re Building

Why We’re Confident

Fiscal Year 2026 Guidance

New Investor Relations Website