CleanSpark's quarterly 10-Q for the period ended March 31, 2026 is the kind of filing that rewards a reader who ignores the headline machine count and looks at the production line. The number that matters for any bitcoin miner is not how much capacity it has installed but how many coins that capacity actually produced and at what price they could be valued or sold. On both counts, CleanSpark's quarter shows the gravitational pull that rising network difficulty and a falling bitcoin price exert on even a large, efficient operator.

The company reports an average operating hashrate of 47.3 exahash per second for the quarter, following a peak of 50 EH/s reached during fiscal 2025. That is genuine top-tier scale among listed miners. But capacity and output are not the same thing, and the production figures tell the real story.

"During the three months ended March 31, 2026, we mined 1,795 bitcoin with an average bitcoin price of $75,989 as compared to 1,957 bitcoin with an average bitcoin price of $92,870 during the three months ended March 31, 2025."— CleanSpark, Inc. Form 10-Q, source

Two things moved against the company year over year, and they compound. First, CleanSpark mined fewer coins — 1,795 versus 1,957, a decline of roughly 8% — despite running substantial hashrate. That is the difficulty effect: as global hashrate grows, each unit of a miner's own capacity captures a smaller share of the fixed daily block reward. A miner can add machines and still mine fewer coins if the rest of the network adds machines faster. Second, the average bitcoin price fell from $92,870 to $75,989, an 18% drop. Fewer coins at a lower price is the double squeeze that defines a bear-ish quarter for the mining sector, and it shows up directly in revenue.

Reading depreciation per coin

The 10-Q is unusually explicit about how it allocates costs, and it draws attention to a metric worth understanding. The filing presents non-cash miner depreciation on a per-bitcoin basis — "calculated by dividing miner depreciation expense in our owned facilities by the number of bitcoin mined in the owned facilities," as the company puts it. This is a more honest disclosure than many miners offer, because it forces the depreciation of the hardware into the cost of each coin produced. The arithmetic is unforgiving: when you mine fewer coins, the same fixed depreciation spreads across a smaller numerator, so depreciation per bitcoin rises even if nothing about the machines changed. That is the quiet cost inflation that a falling-output quarter bakes in.

For investors, the per-coin framing is the right lens. Energy is the variable cost everyone talks about, but for a capital-intensive miner the depreciation of the fleet is a large fixed cost that does not disappear when the price falls. A miner can have cheap power and still lose money if hardware depreciation per coin climbs above the price at which it can sell what it mines. The 10-Q's choice to spell out the per-bitcoin depreciation calculation is a signal that CleanSpark wants investors to evaluate it on a fully loaded basis rather than on a power-only "cost to mine" figure.

What 47 EH/s buys in a hard market

The broader point this filing illustrates is structural to the entire sector. Mining is a commodity business where the product price is set by a global market and the cost of production rises mechanically as competitors add capacity. A miner's only durable advantages are the cost of its electricity, the efficiency of its machines, and its discipline about when to sell. CleanSpark's 47.3 EH/s puts it among the largest, which helps with economies of scale, but scale does not insulate it from the two forces that defined this quarter: network difficulty eating into per-machine yield, and a price decline shrinking the value of every coin produced.

The filing also references the company's performance-based equity, noting that as of March 31, 2026 it estimated 100% of an operational PSU tranche would vest based on locations actively marketed or expected to be marketed with the power capability to meet performance targets by March 2027 — a reminder that part of management's compensation is tied to bringing energized capacity online. Growth in capacity is the lever CleanSpark is pulling to offset the difficulty headwind. Whether it works depends on whether new energized hashrate comes on faster than the network's, and whether the bitcoin price recovers enough to make the marginal coin worth mining.

It is also worth situating this quarter against the longer arc of mining economics. Every bitcoin halving cuts the block subsidy in half, mechanically halving the largest component of miner revenue overnight while costs stay roughly constant. Between halvings, the network's difficulty adjustment steadily raises the bar as more hashrate comes online, so a miner that stands still loses ground continuously. The only durable responses are to lower the cost per terahash — through newer, more efficient machines and cheaper power — or to diversify revenue beyond the block reward, which is why so many miners have pushed into transaction-fee capture, energy arbitrage, and even AI/high-performance-computing hosting. CleanSpark's filing is a snapshot of a company running hard inside that treadmill: substantial hashrate, transparent per-coin accounting, and a production line that nonetheless yielded fewer coins at a lower price than a year before. The structural pressure does not relent between halvings; it intensifies, and a single soft-price quarter makes that pressure visible in a way a bull market hides.

The takeaway for anyone following the miners is simple and was visible across the whole cohort this quarter: against a bitcoin price in the mid-$70,000s, even a 47 EH/s operator mined fewer coins for less money than it did a year earlier. CleanSpark's relatively transparent per-coin cost disclosure makes it a useful benchmark for the rest of the sector — and the benchmark this quarter is a market that is squeezing producers from both the output side and the price side at once.