CleanSpark's $6.6B Lease May Save CLSK From a Bitcoin Squeeze

CleanSpark is being judged on two businesses at once

CleanSpark's $6.6 billion lease may be the catalyst that helps CLSK weather a tougher Bitcoin cycle.

The latest earnings call highlighted the company's split identity: a mining business already feeling pressure today, and an infrastructure platform that investors may not fully value yet.

Why the bear case still starts with the quarter

CleanSpark's recent mining quarter was weak. The company reported Q1 2026 revenue of $181 million, down 19% from the prior quarter, while gross margin fell to 47% from 57% a year earlier. Reported net loss was about $379 million, mainly because of noncash Bitcoin mark-to-market adjustments, and normalized EBITDA was $55 million. That supports the bear argument that current mining economics are under pressure.

The bull case is broader than one soft quarter. CleanSpark still held more than 13,000 Bitcoin at quarter-end, with a value of about $1.15 billion, so shareholders remain exposed to Bitcoin through both operations and the balance sheet.

Sandersville is the key test of CleanSpark's pivot

CleanSpark has now secured a 20-year triple-net lease at Sandersville that includes 175 MW of critical IT load and $6.6 billion in contracted revenue over the initial term. Management expects deliveries to begin in Q4 2027 and has outlined roughly $330 million in average annual NOI with a nearly 100% NOI contribution margin. If that project executes, it could materially change what kind of business investors think CleanSpark is becoming.

Mining economics vs. infrastructure-style cash flow

This is really a valuation debate. CleanSpark's recent quarter still reflected pressure from rising Bitcoin network difficulty and higher power prices, which keeps the old miner narrative alive. But the market is now being asked to evaluate whether Sandersville can shift CleanSpark toward more stable, infrastructure-style cash flows.

The lease structure matters more than the headline number. A 20-year triple-net lease with a high investment-grade tenant would move earnings away from mining volatility and closer to long-duration, customer-funded income. Management's expectation of a nearly 100% NOI margin and about $330 million in average annual NOI supports that argument. If the market starts to see that stream as credible, CLSK could be valued on earnings quality rather than only on Bitcoin price.

Funding matters as much as demand

Investors do not need the build-out to be complete to see a rerating. They need proof that demand is contracted, the tenant is creditworthy, and the capital structure limits dilution risk. On that front, bulls can point to CleanSpark's statement that its equity contribution is fully funded and that most of the estimated $1.75 billion to $2.1 billion project cost is expected to be financed through project debt. If that remains the case, shareholders have a better chance of capturing upside from multiple expansion rather than absorbing dilution.

Texas remains the main execution risk

One strong lease does not erase the need for the core mining business to remain viable while Sandersville comes online. It also does not eliminate regulatory risk in Texas, where CleanSpark still needs clarity on final ERCOT and PUCT determinations.

That uncertainty is the reason Texas still cannot be underwritten like a booked platform. The company has said timelines and outcomes beyond the relevant regulatory hearings remain unclear, so investors should watch whether CleanSpark can keep converting power, land, and regulatory progress into firm demand.

What would change the debate from here?

• Mining stability: less pressure on revenue, margin, and normalized EBITDA would give the company more room while the infrastructure build-out advances.
• Sandersville progress: confirmed construction milestones and an on-track first-delivery date would make the lease more than a headline.
• Texas clarity: defined regulatory outcomes would show that CleanSpark can expand the platform beyond Georgia.