<div><div>QumulusAI filed an S-1 for a direct listing on the Nasdaq Global Market under the ticker “QMLS.” The company will not raise capital in the listing; instead, registered shareholders may resell their shares after effectiveness, with most insider shares subject to a six‑month restriction. Chardan Capital Markets is serving as financial advisor for the listing process.Business DescriptionQumulusAI is a cloud infrastructure company focused on delivering GPU‑powered compute for artificial intelligence and high‑performance computing workloads. The platform provides both on‑demand and reserved access to NVIDIA‑based clusters for training and inference, delivered through direct enterprise engagements and marketplace partners. The company differentiates by activating new GPU capacity in about 90 days, using a mix of colocation sites and owned power‑intensive assets to accelerate time to revenue and maintain cost control.Revenue today comes from three areas: compute power (GPUaaS), hosting services for third‑party mining customers, and bitcoin self‑mining (now a smaller portion as the company pivots to HPC). QumulusAI reaches thousands of developers through marketplace partners like RunPod (80/20 revenue share), while its direct sales channel supports larger, customized enterprise deployments. The company emphasizes end‑to‑end control of the stack—data center, networking, cooling, and power procurement—to improve reliability, utilization, and economics versus general‑purpose cloud options.Market Overview<ul><li>Total addressable market: Global AI infrastructure spend projected to approach $337 billion in 2025 and reach $749 billion by 2028.</li><li>Market growth: Generative AI could add $2.6 trillion to $4.4 trillion in annual productivity; broader AI potential estimated at $15.5 trillion to $22.9 trillion annually by 2040.</li><li>Market position: QumulusAI targets underserved small and mid‑market AI users while supporting enterprise projects, with rapid 90‑day GPU deployments.</li><li>Key competitors: Hyperscalers (AWS, Azure, Google Cloud, IBM, Oracle) and AI‑focused clouds (CoreWeave, Lambda, Crusoe).</li><li>Industry trends: Rising demand for GPU capacity, supply chain constraints, power availability challenges, rapid AI adoption, and evolving AI regulation.</li></ul>Operational Metrics<ul><li>Customers: Access to a marketplace base of 10,000+ AI developers via RunPod; enterprise customers served directly.</li><li>Locations: HQ in Marietta, Georgia; colocation in Marietta, GA and Kansas City, MO; power sites in Watonga and Tulsa, Oklahoma; Denton, Texas under development.</li><li>Geographic presence: U.S. footprint with multi‑site deployment and expansion capacity.</li><li>Partnerships: RunPod (channel), hosted.ai (orchestration/virtualization), colocation partners CoreSite, H5, and NOCIX.</li><li>Installed GPUs: ~800 GPUs active as of Dec. 1, 2025; ~1,100 including in‑process deliveries.</li><li>Power capacity: ~60 MW active grid power with &gt;40 MW additional available; rights of first refusal for 30 MW of IT load and plans exceeding 120 MW.</li><li>Deployment speed: New GPU infrastructure activated and revenue‑producing in ~90 days.</li><li>Channel mix: Marketplace revenue share typically 80% QumulusAI / 20% partner; pursuing Preferred Provider Program.</li><li>2026 plan: Expect ~11 MW of HPC IT load (~15.4 MW total power) and a greater share of power allocated to HPC vs. blockchain.</li></ul>Financials Highlights<ul><li>Revenue (current): $8,057,476 (nine months ended Sept. 30, 2025)</li><li>Revenue growth: Up ~31% YoY for the nine months ended Sept. 30, 2025</li><li>Operating income: Operating loss of $(5,866,060) (nine months ended Sept. 30, 2025)</li><li>Net income: $831,828 net income (nine months ended Sept. 30, 2025); FY2024 net loss $(13,184,374)</li></ul>Management<ul><li>Michael Maniscalco, Chief Executive Officer - Former CTO of Applied Digital; two decades in AI/HPC, IoT, and infrastructure; Georgia Tech B.S. in Computer Science.</li><li>Scott Krosnowski, Chief Financial Officer - 25+ years in finance leadership roles; prior SVP Finance at SBI Growth Advisory and CFO roles; American University B.S.B.A.</li><li>Ankur Chatterjee, Chief Integration Officer - 20+ years in infrastructure and systems integration; led data center and power solutions; Georgia Tech B.S. in Management (IS minor).</li><li>Ryan DiRocco, Chief Technology Officer - Former CTO at Performive; 20+ years in virtualization, networking, and managed services.</li><li>Stephen Hunton, Chief Marketing Officer - Ex‑IBM VP of Global Social &amp; Content; 20+ years in integrated digital marketing for enterprise tech brands.</li></ul>IPO Structure<ul><li>Issuer: QumulusAI, Inc.</li><li>Filing date: December 31, 2025</li><li>Proposed ticker: QMLS</li><li>Exchange: Nasdaq Global Market</li><li>Lead underwriters: No underwriters; Chardan Capital Markets LLC as financial advisor</li><li>Use of proceeds: QumulusAI will not receive proceeds; any proceeds go to selling shareholders.</li></ul>Original SEC Filing: QumulusAI, Inc. - S-1 - Dec. 31, 2025<section>Disclaimer<div>This is an AI-powered summary. It may contain inaccuracies. Consider verifying important information with the source. Please note this summary is solely based on documents filed with the SEC.</div></section></div></div>

QumulusAI has filed an S-1 for a direct listing on the Nasdaq Global Market under the ticker "QMLS." The company will not raise capital but allows registered shareholders to resell shares post-effectiveness. QumulusAI specializes in GPU-powered cloud infrastructure for AI and high-performance computing, targeting small and mid-market users. The global AI infrastructure market is projected to reach $749 billion by 2028. For the nine months ending Sept. 30, 2025, QumulusAI reported $8.06 million in revenue, a 31% YoY increase, but an operating loss of $5.87 million.

QumulusAI, AI GPU Cloud Infrastructure Provider, Files for Nasdaq Global Market IPO