$QMLS reports tomorrow. $246M in agreements. $194M market cap. Show us the racks.

Tomorrow is the first time QumulusAI has to sit down in public and explain the quarter.

Quick catalyst update to yesterday's GPU-landlord DD.

The company reports Q2 results after the close on Tuesday, August 25, followed by its \[first public earnings call at 5 p.m. ET\](https://www.qumulusai.com/articles/qumulusai-to-report-second-quarter-2026-financial-results-on-august-25-2026).

QMLS traded at $5.89 after hours on Monday. Using the 32.87 million basic shares in the \[final prospectus\](https://www.sec.gov/Archives/edgar/data/2084026/000143774926023622/quma20260714\_424b4.htm), the entire company is valued near $193.6 million.

QumulusAI has announced more than $246 million in multi-year customer agreements since June.

That gap is why tomorrow matters.

\## Keep the calendar straight

Q2 ended June 30. The $18 million take-or-pay agreement, the $32 million B300 agreement, the $71.9 million agreement, the 1,632-GPU B300 purchase, the DRW deal, and the new Atlanta capacity agreement all came later.

The $246 million will not teleport into tomorrow's revenue line. Anyone expecting that should give the spreadsheet back to its legal guardian.

What Q2 can show is whether the original business was already accelerating before the July news started flying. Q1 revenue was $3.42 million, gross profit was $1.28 million, and operating loss was $5.53 million. That is the starting line.

\## The number I care about is six

On August 6, management said it was \[delivering six customer deployments\](https://www.qumulusai.com/articles/qumulusai-to-participate-in-needham-virtual-ai-infrastructure-1x1-conference), with several close to completion and the rest expected to finish this quarter.

Tomorrow I want names where they can give them, GPU counts, energized megawatts, go-live dates, and the amount of revenue those deployments can produce. Contracts are nice. Powered racks sending invoices are nicer.

QumulusAI announced \[$124.4 million of three-year agreements\](https://www.qumulusai.com/articles/qumulusal-signs-more-than-124-million-in-ai-inference-infrastructure-agreements) before Q2 closed, including nearly $21.9 million in upfront customer commitments. If management can connect that cash and contracted demand to actual deployment progress, the bear argument gets much harder.

Right now the market appears to value every press release at roughly zero until a server starts humming. Fair enough. Tomorrow is management's chance to introduce the market to electricity.

\## The post-quarter setup got stronger

Since June, QumulusAI has announced more than $246 million across four two- and three-year agreement blocks. Simple division by the stated terms produces more than $90 million a year of potential contract value before accounting for deployment timing, revenue recognition, margins, financing, or customer performance.

The company then added \[DRW as a B300 customer\](https://www.qumulusai.com/articles/qumulusai-signs-gpu-as-a-service-agreement-with-drw-for-nvidia-blackwell-b300-capacity-annually-renewable-up-to-four-years), signed an \[agentic hedge fund\](https://www.qumulusai.com/articles/qumulusai-signs-agreement-with-agentic-hedge-fund-to-provide-nvidia-blackwell-gpu-capacity) at market-rate compute pricing plus a share of trading profits, and secured \[3.75 MW of Atlanta capacity\](https://www.qumulusai.com/articles/qumulusai-anchors-metro-atlanta-data-center-site-with-a-contracted-375-mw-and-a-path-to-1075-mw) that could support up to 2,048 B300-class GPUs once the site is delivered. The DRW and hedge-fund agreement values were not disclosed, and the Atlanta delivery remains conditional.

That is demand, hardware, power, and deployment all moving in the same direction. The income statement now has to catch up.

\## The $300 million question

Management guided to \[$300 million of forward ARR and 18 MW of capacity by December 31\](https://www.qumulusai.com/articles/qumulusai-issues-fiscal-year-2026-guidance-300-million-in-forward-arr-backed-by-18-mw-of-capacity).

At a $193.6 million basic market cap, the stock trades at roughly 0.65 times that target.

Adult supervision remains required. QumulusAI's forward-ARR definition includes executed contract revenue, expected renewals, deposit-backed reservations, and projected contract signings. The definition reaches far beyond GAAP revenue.

Tomorrow I want the bridge. How much is executed? How much is deposit-backed? How much is still projected? How much capacity is active today? What portion should begin generating revenue during Q3 and Q4?

A clean bridge could change how the market values the whole story.

At $193.6 million, perfection is nowhere in the quote. QMLS only needs to prove that a meaningful slice of the announced demand now has a power cord. A clean deployment update would give the market a reason to stop treating this company like a PowerPoint with a utility bill.

That is why I am bullish into the call.

\## The bear case gets the microphone too

The company entered Q2 with $16.1 million of cash, $102.4 million of liabilities, material weaknesses in financial controls, major capital requirements, and heavy RunPod concentration. The direct listing raised no fresh primary capital. The Blackwell buildout relies on leases, convertible notes, customer deposits, and equipment financing.

If revenue stays flat, deployments slip, guidance gets slippery, or management avoids the financing question, the bears win this round.

If revenue is moving, margins hold, the six deployments are real, and management can explain how funded hardware becomes billed capacity, $5.89 starts looking like the market read the first page and went home.

That is the setup. Tomorrow QMLS opens the books. I am watching for racks, revenue, and receipts.

High-risk small-cap thesis. Verify the filings and size your own risk.