Measuring the true cost for miners is no simple task. It involves at least three layers: cash costs (electricity, operations, labor), depreciation (mining equipment amortization), and total costs (including financial expenses and capital expenditures). Depending on the calculation method, the difference can reach tens of thousands of dollars.
According to CoinShares’ Q1 2026 mining report, the weighted average cash cost for public mining companies to produce one Bitcoin had already climbed to about $79,995 by Q4 2025. When factoring in depreciation, administrative expenses, and capital expenditures, some estimates put the total cost in the $83,000 to $90,000 range.
As of April 8, 2026, Gate market data shows Bitcoin spot prices consolidating near $71,000. This means that for each Bitcoin mined, companies are facing a theoretical loss of $12,000 to $19,000. Roughly 15% to 20% of mining machines worldwide are now operating at a loss.
Multiple forces are driving these soaring costs: the April 2024 Bitcoin halving cut block rewards from 6.25 BTC to 3.125 BTC, slashing miners’ core revenue in half. Global energy prices continue to rise. The rollout of next-generation ASIC miners has pushed up capital expenditures. Network difficulty hit a record high of 156 T by the end of 2025. Together, these factors have transformed mining from a "money printer" into a "cash burner."
How Large Is the Scale of Miner Sell-Offs?
With cash losses of nearly $20,000 per coin, mining companies’ balance sheets are under unprecedented pressure. In Q1 2026, public miners collectively launched a wave of large-scale Bitcoin sales.
Riot Platforms sold 3,778 Bitcoins during the quarter at an average price of $76,626 each, generating about $289.5 million in revenue. This sale volume was roughly 2.6 times their production for the quarter (1,473 BTC). Riot’s Bitcoin holdings dropped from around 19,223 to 15,680, an 18% decrease year-over-year.
MARA Holdings’ sell-off was even more significant. Between March 4 and March 25, the company sold 15,133 Bitcoins in stages, raising about $1.1 billion—mainly to repurchase convertible bonds. This reduction far exceeded the "sell only newly mined coins" approach—MARA mined only 8,799 BTC in all of 2025, while its March sales alone were nearly 1.7 times its annual output.
The sell-off wasn’t limited to the largest miners. Bitfarms announced it would liquidate all its Bitcoin reserves to redeploy funds into building AI data centers. Core Scientific stated it planned to sell its entire Bitcoin holdings—about 2,500 coins—in Q1 2026. Altogether, several public miners sold over 19,000 BTC in Q1 2026.
This selling pressure has spread to the network level. Bitcoin hashprice dropped from about $63/PH/s/day in July 2025 to just $28–$30/PH/s/day by early March 2026, marking the lowest point since the halving.
What’s Changing in Post-Halving Supply Economics?
The wave of selling reflects a structural reshaping of supply after the halving.
The April 2024 halving reduced block rewards from 6.25 BTC to 3.125 BTC, cutting daily new Bitcoin issuance from about 900 to 450. Traditionally, such supply contractions should boost prices. However, the reality from 2025 to 2026 diverged from this narrative: Bitcoin’s price fell more than 45% from its $126,000 peak in October 2025 to the $65,000 range.
The core reason is that the halving’s supply shock was greatly diluted by ETF and institutional capital inflows. In the past, halvings meant a drop in daily marginal sell pressure by several hundred BTC. Today, however, institutional holdings far outweigh daily miner output. Miners’ sales no longer dominate the market, but their structural impact is more profound.
On the hashrate side, Bitcoin network hashrate peaked at about 1,160 EH/s at the end of 2025, but Q1 2026 saw the first quarterly decline in six years—a drop of about 4% to 10% since the start of the year, now stabilizing between 900 and 1,020 EH/s. At the same time, mining difficulty saw a sharp 7.76% downward adjustment in late March, the largest drop in over a year.
Together, these indicators point to one signal: inefficient miners are being systematically squeezed out. As noted by institutions like CICC and GF Futures, the steepening miner cost curve signals a "capacity clearing" process, paving the way for future supply structure optimization. The supply contraction effect of the halving hasn’t disappeared—it’s just being masked by the price downturn. Once prices rebound, lost hashrate will be slow to return, and supply elasticity may be lower than historical averages.
What Three Paths Do Miners Face?
With cash losses of nearly $20,000 per coin, miners have very limited options. As seen from industry trends in Q1 2026, there are three main paths forward.
Path 1: Sell Reserves for Cash. This is currently the most common choice. Riot sold 3,778 BTC for about $289.5 million, mainly to cover the construction and operational costs of the Corsicana AI data center. MARA sold 15,133 BTC for about $1.1 billion, reducing its outstanding debt from $3.3 billion to $2.3 billion—a 30% drop—and saving about $88.1 million in interest payments. The trade-off is giving up potential future Bitcoin price gains in exchange for immediate survival.
Path 2: Pivot to AI and High-Performance Computing (HPC). Several public miners have announced over $7 billion in AI and HPC contracts. Hut 8 signed a 15-year, $700 million lease agreement with Google. Riot is repurposing two-thirds of its Corsicana facility for AI workloads. AI hosting offers predictable, high-margin revenue—typically 80% to 90% operating profit—via multi-year, dollar-denominated contracts, a sharp contrast to the volatility of Bitcoin block rewards. For some leading firms, AI-related revenue already accounts for about 30% of total income and could reach as high as 70% by the end of 2026.
Path 3: Shut Down Inefficient Miners. This is the most direct, but also the most disruptive, option. About 252 EH/s of marginal hashrate is currently offline globally, mainly outdated hardware with efficiency below the 25 J/TH standard. This route reduces competition and improves profitability for surviving miners, but means a complete market exit for those who shut down.
These paths aren’t mutually exclusive. Riot and MARA have adopted a "sell-off + pivot" strategy, while Core Scientific and Bitfarms are leaning toward "liquidation + full transformation." Overall, the AI pivot is the most strategically valuable direction—it enables miners to shift from passive price takers to active providers of compute infrastructure.
Why Has "Hodling" Collapsed Systematically in 2026?
The "hodling" philosophy among Bitcoin miners—holding onto mined coins long-term instead of selling immediately—has been almost a matter of faith in the industry. The core logic: miners’ costs are denominated in fiat, but output is in Bitcoin. If you believe in Bitcoin’s long-term appreciation, delaying sales locks in higher fiat returns. This strategy was validated on a large scale during the 2020–2021 bull market.
But in 2026, this belief is unraveling. Data shows that public miners sold over 15,000 BTC between October 2025 and February 2026, marking the most intense treasury liquidation in this cycle. In early 2026, MARA revised its reserve policy from only selling newly mined coins to allowing discretionary sales of all balance sheet holdings. Core Scientific stated it plans to liquidate the vast majority of its Bitcoin reserves.
There are deep reasons for this shift. First, mining margins have turned negative—holding newly mined BTC means incurring about $19,000 in cash losses per coin, an unsustainable burden for any balance sheet. Second, miners’ financing structures have changed: pressure from maturing convertible bonds and capital needs for AI pivots are forcing management to treat Bitcoin as a liquid asset, not a long-term reserve. Third, Bitcoin ETFs have changed the market structure—institutions can now gain exposure through ETFs, so miners’ holdings no longer command a scarcity premium.
Importantly, miner sell-offs don’t mean a total industry retreat. The market is structurally splitting: Strategy bought 44,377 BTC in March alone, and Japanese public company Metaplanet added 5,075 BTC in Q1. Bitcoin demand hasn’t disappeared; it’s just concentrating among stronger players.
Can AI Transformation Provide a Long-Term Path for Miners?
Pivoting to AI is the most significant structural change in Bitcoin mining in 2026. But is this truly a long-term solution, or just a temporary safe haven?
From a revenue perspective, the logic for AI transformation holds. Bitcoin mining income is highly volatile, tied to coin prices and network fees. In contrast, AI hosting contracts are multi-year, dollar-denominated agreements that provide predictable cash flow. CoinShares estimates AI hosting can deliver operating margins of 80% to 90%, far exceeding current Bitcoin mining margins. By the end of 2026, some leading miners could see up to 70% of their revenue from AI, effectively becoming data center operators.
On the capital allocation side, the AI shift is accelerating. Bitdeer disclosed its GPU cloud business data: 2,096 GPUs deployed, 64% utilization, and about $21 million in annual recurring revenue. Core Scientific secured a $500 million credit line from Morgan Stanley to fund expansion into HPC hosting.
But the AI pivot isn’t without limits. The transition is extremely costly—converting a Bitcoin mine to an AI data center can run $1 million to $1.5 million per megawatt, which is why only well-capitalized players can make the shift. In addition, AI data centers have much higher requirements for power stability, network latency, and cooling than Bitcoin mining—not all sites are suitable for conversion.
A deeper question remains: as more miners move into AI infrastructure, could the AI compute market become oversupplied? For now, hyperscale cloud providers like Google are still rapidly expanding their demand for compute, so miners pivoting to AI are filling a supply gap. As long as this gap exists, the AI transformation offers miners a sustainable path forward.
Conclusion
The Bitcoin mining industry is facing its toughest profitability crisis since the 2024 halving. Public miners’ weighted average cash cost is about $79,995, while spot BTC trades between $68,000 and $70,000—implying a theoretical loss of around $19,000 per coin. In Q1 2026, Riot sold 3,778 BTC (2.6 times its quarterly production), MARA sold 15,133 BTC (about $1.1 billion), and several public miners together sold over 19,000 BTC. Bitcoin network hashrate saw its first quarterly decline in six years, with about 252 EH/s of marginal hashrate offline.
Miners face three main responses: selling reserves for cash, pivoting to AI and HPC, or shutting down inefficient machines. Among these, the AI pivot is the most strategically valuable—public miners have announced over $7 billion in AI and HPC contracts, and some expect up to 70% of revenue from AI by the end of 2026. The "hodling" philosophy is collapsing, and the market is splitting structurally—miners are selling due to operational pressure, while institutions like Strategy continue to accumulate. The supply contraction effect of the halving hasn’t disappeared, but is masked by the price downturn. Once prices rebound, lost hashrate will be slow to return, and supply elasticity may be lower than historical averages.
FAQ
Q: Why has the cost of mining Bitcoin surpassed $80,000?
Mining costs are mainly driven by electricity expenses, miner depreciation, operational costs, and financial expenses. After the April 2024 halving, block rewards dropped from 6.25 BTC to 3.125 BTC, cutting miners’ unit revenue in half. At the same time, rising global energy prices, next-generation ASIC miner upgrades, and network difficulty reaching a record 156 T by the end of 2025 have all pushed total costs higher. According to CoinShares’ Q1 2026 mining report, public miners’ weighted average cash cost has surged to about $79,995.
Q: How does miner selling affect Bitcoin market supply?
Miners are one of the most stable sources of Bitcoin supply. In Q1 2026, Riot sold 3,778 BTC, MARA sold 15,133 BTC, and several public miners together sold over 19,000 BTC. These sales increase short-term supply pressure, but also mean that future selling pressure will be limited—since the remaining miners are mainly those with access to low-cost energy and advanced ASIC infrastructure. On the hashrate side, as inefficient miners are squeezed out, the industry’s supply structure becomes more optimized.
Q: Does declining hashrate mean Bitcoin network security is at risk?
Bitcoin’s network has a difficulty adjustment mechanism that automatically adapts to changes in hashrate. The current decline is mainly due to profitability pressures forcing inefficient miners offline, not network security issues. Global hashrate remains stable between 900 and 1,020 EH/s, far above historical levels. Network difficulty was lowered by about 7.76% in late March, improving profitability for surviving miners and demonstrating Bitcoin’s self-adjusting protocol.
Q: Is AI transformation a viable path for all miners?
AI transformation isn’t accessible to every miner. The transition is expensive—converting a Bitcoin mine to an AI data center can cost $1 million to $1.5 million per megawatt, and requires much higher standards for power stability, network latency, and cooling. Only well-capitalized public miners can make this shift. For small and mid-sized miners, shutting down or selling machines remains the practical choice.
