Analyst(s): Brendan Burke
Publication Date: August 28, 2026
QumulusAI’s Q2 FY 2026 earnings report points to a business moving from marketplace dependence toward direct AI compute relationships. The quarter centered on new take-or-pay contracts, GPU fleet expansion, and early public-company positioning in AI infrastructure.
What Is Covered in This Article:
- QumulusAI’s Q2 FY 2026 financial results
- Direct customer AI compute momentum
- GPU fleet and capacity expansion
- NVIDIA partnership and supply positioning
- Outlook and Final Thoughts
The News: QumulusAI (NASDAQ: QMLS) reported Q2 FY 2026 revenue of $6.713 million, up 117.6% year over year (YoY), above Wall Street consensus of $3.230 million. AI Compute revenue was approximately $5.6 million, representing 84% of total revenue, compared with $1.3 million, or 43%, in Q2 FY 2025. Gross profit was $4.470 million, and gross margin was 66.6%, up from 55.1% a year earlier. Operating loss was $7.671 million, compared with an operating loss of $2.182 million in Q2 FY 2025. Net loss was $22.776 million, compared with net income of $12.119 million a year earlier. Diluted net loss per share stood at $0.72, compared with diluted earnings per share of $0.46 in Q2 FY 2025.
“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.”
QumulusAI Q2 FY 2026: 118% Revenue Growth for Hyperspeed AI Compute Deployment
Analyst Take: QumulusAI’s Q2 FY 2026 results show a shift toward direct customer contracts and away from dependence on a single marketplace. That transition gives QumulusAI more control over customer economics, capacity allocation, and recurring revenue quality. The challenge is now execution, because GPU procurement, power availability, financing, and colocation access all need to scale at the same pace as demand.
Direct Contracts Shift the Revenue Model
QumulusAI signed 21 new direct customer AI compute contracts in Q2 FY 2026 with an aggregate expected take-or-pay contract value of $169.7 million. Revenue from direct customer relationships accounted for more than 96% of the recurring revenue base at quarter-end, up from less than 10% a year earlier. That shift reduces reliance on a single marketplace and gives the company a clearer path to pricing control. It also changes the commercial profile of the business, because take-or-pay structures can improve visibility if capacity comes online on schedule. The reported $120+ million in new customer agreements after the quarter, including a three-year agreement valued at above $71 million, points to continued demand beyond the reported period. QumulusAI’s next phase depends less on finding demand and more on converting contracted demand into deployed revenue.
GPU Capacity Becomes the Operating Constraint
QumulusAI grew its deployed GPU fleet 224% from 952 to 3,088 during the quarter. That expansion gives the company more capacity to serve AI compute customers, but it also increases operational complexity. The company ended Q2 FY 2026 with 8 megawatts of high-performance computing capacity under executed lease and colocation agreements. The Atlanta colocation agreement adds up to 3.75 megawatts, with a right of first offer on up to 7 megawatts of expansion capacity at the same site. The purchase of 1,632 NVIDIA Blackwell B300 GPUs gives QumulusAI a newer supply base for customers seeking higher-performance AI infrastructure. The implementation of the hyperspeed deployment model will support the contract base converting into recognized revenue in months rather than pushing to next year.
NVIDIA Partnership Raises Execution Requirements
QumulusAI became an NVIDIA Cloud Partner on July 17, 2026, shortly after beginning trading on the Nasdaq Global Market. The partnership should improve credibility with AI compute buyers who prioritize GPU availability and deployment reliability. It also places more pressure on QumulusAI to meet operating standards tied to reference designs, software stack readiness, and customer support. The GPU-as-a-Service agreement with DRW shows the company is reaching customers with demanding compute needs and sophisticated GPU pricing insights, given the customer’s incubation of the Silicon Data business. The Blackwell B300 purchase also links QumulusAI’s near-term roadmap to the availability and deployment of advanced NVIDIA systems. The NVIDIA relationship can open doors, but QumulusAI still has to prove it can deliver scale with consistent service economics.
Outlook and Final Thoughts
QumulusAI did not reaffirm its previously issued guidance, including its $300 million ARR target and plan to reach 18 megawatts of HPC capacity by year-end. The lack of an updated financial outlook increases the importance of execution milestones during H2 FY 2026, particularly the pace at which contracted capacity becomes operational and begins generating revenue. Management has made clear that customer demand is not the primary constraint, shifting attention toward infrastructure deployment, financing, and the economics of scaling the platform.
The key question is whether QumulusAI can translate its rapidly expanding contract base into revenue without capital requirements becoming a limiting factor. Take-or-pay structures provide greater visibility, but revenue realization still depends on GPUs, power, networking, and colocation capacity becoming available on schedule. Financing costs and customer concentration also become more important as the company scales from a relatively small revenue base. If QumulusAI can maintain deployment speed while improving capital efficiency, then the shift toward direct AI compute contracts could support a more predictable and scalable revenue model.
See the full press release on QumulusAI’s Q2 FY 2026 financial results on the company website.
Declaration of generative AI and AI-assisted technologies in the writing process: This content has been generated with the support of artificial intelligence technologies. Due to the fast pace of content creation and the continuous evolution of data and information, The Futurum Group and its analysts strive to ensure the accuracy and factual integrity of the information presented. However, the opinions and interpretations expressed in this content reflect those of the individual author/analyst. The Futurum Group makes no guarantees regarding the completeness, accuracy, or reliability of any information contained herein. Readers are encouraged to verify facts independently and consult relevant sources for further clarification.
Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.
Read the full Futurum Group Disclosure.
Other Insights From Futurum:
How Intel 18A-P Transistor Breakthroughs Shape Diamond Rapids CPU Design
NVIDIA’s Credit Support Buys Exclusivity at OpenAI’s Ohio Data Center
Micron Research Labs: Can $1 Billion a Year Out-Research Samsung and SK hynix?
Author Information
Brendan is Research Director, Semiconductors, Supply Chain, and Emerging Tech. He advises clients on strategic initiatives and leads the Futurum Semiconductors Practice. He is an experienced tech industry analyst who has guided tech leaders in identifying market opportunities spanning edge processors, generative AI applications, and hyperscale data centers.
Before joining Futurum, Brendan consulted with global AI leaders and served as a Senior Analyst in Emerging Technology Research at PitchBook. At PitchBook, he developed market intelligence tools for AI, highlighted by one of the industry’s most comprehensive AI semiconductor market landscapes encompassing both public and private companies. He has advised Fortune 100 tech giants, growth-stage innovators, global investors, and leading market research firms. Before PitchBook, he led research teams in tech investment banking and market research.
Brendan is based in Seattle, Washington. He has a Bachelor of Arts Degree from Amherst College.
