Crypto Mining Trends in 2026: Margins, AI and Energy

Crypto mining trends in 2026: record-low hashprice, falling hashrate, the pivot to AI computing, grid constraints, and how to read miner profitability.

Cryptela EditorialNov 9, 2026

Crypto Mining Trends in 2026: Margins, AI and Energy

The defining crypto mining trend of 2026 is a squeeze on margins. Bitcoin’s price fell, hashprice (the daily revenue a miner earns per unit of computing power) hit record lows, and many miners are redirecting power and capital toward AI computing. Mining is still active and still profitable for operators with cheap electricity and efficient machines. For almost everyone else, it is not.

This guide explains how mining works in plain terms, walks through the main trends with dated figures, and shows how to judge profitability yourself. It is educational content, not financial advice.

Cryptela

cryptela.com

Track the Markets Behind Mining

Explore real-time cryptocurrency prices, trading volumes, historical performance, and news to follow the market forces shaping mining margins.

Explore crypto markets

Key Takeaways

Hashprice fell to an all-time monthly low of US$27.7 per petahash per day in June 2026, according to CoinShares’ Q2 2026 mining report, and the listed miners were below cash breakeven in aggregate.
Bitcoin’s network hashrate has dropped from its October 2025 peak, and public miners are shifting power toward AI and high-performance computing (HPC).
Electricity price and machine efficiency decide who profits. At 8¢/kWh only some models earn more than their power bill, and at 12¢/kWh none do, per one November 2026 dataset.
Energised, grid-connected sites are becoming more valuable, so power access matters as much as hardware.
DeFi is a separate market with its own indicators, such as total value locked. It is not a mining metric.

How crypto mining works in brief
Trend 1: The margin squeeze and record-low hashprice
Trend 2: Hashrate falls and difficulty adjusts downward
Trend 3: Miners pivot to AI and high-performance computing
Trend 4: Power access becomes the scarce asset
Trend 5: Efficiency is the dividing line
Is crypto mining still profitable?
Fees, halvings and the long-run revenue model
Cloud mining and pools: what readers should know
Where DeFi fits, and how it differs from mining
What to watch next
Reaching crypto readers with your mining news
FAQ
Sources

How crypto mining works in brief

Mining is how proof-of-work blockchains such as Bitcoin confirm transactions and issue new coins. Miners run specialized computers that repeatedly compute SHA-256 hashes, a type of cryptographic fingerprint, until one produces a valid result. The miner that finds it can add the next block and collect the block reward plus transaction fees.

Two features explain most of what follows. First, a miner’s chance of winning a block is proportional to its share of the network’s total hash rate, as the academic paper Dynamics of Bitcoin mining shows. A miner with ten times the hash rate earns ten times more on average, so scale brings no efficiency discount in the protocol itself. Second, the network recalculates difficulty every 2,016 blocks, roughly every two weeks, to keep the average block time near 10 minutes, as Lightspark’s glossary describes.

The same paper reduces profitability to one inequality: revenue (coin price times block reward plus fees) must be at least the cost of electricity and hardware depreciation. Every trend below pushes on one side of that inequality.

Trend 1: The margin squeeze and record-low hashprice

Hashprice is the most useful single number for a miner. It tells you what one petahash per second of computing power earns in a day, before any costs.

The recent path has been steep:

In Q1 2026, CoinShares reported hashprice near US$36 to US$38 per PH/s per day, close to breakeven for many miners, and three consecutive negative difficulty adjustments, the first such streak since July 2022.
ABC Money reported that profitability reached a five-year low in February 2026, with hashprice near $28 and an estimated 15% to 20% of the global fleet operating below breakeven.
In Q2 2026, CoinShares recorded the monthly average hashprice at US$27.7 in June and said Bitcoin ended the quarter at US$58,400, less than half its October 2025 all-time high.

The cause is simple arithmetic. Revenue is priced in Bitcoin, while electricity and hardware are paid for in dollars. When the coin price falls and the block reward stays fixed (it was cut to 3.125 BTC by the April 2024 halving), margins shrink fast.

Prices recover and fall in cycles, so treat these figures as a snapshot of 2026 conditions, not a permanent state.

Trend 2: Hashrate falls and difficulty adjusts downward

When mining stops paying, the least efficient machines switch off. That shows up as falling hashrate, the total computing power securing the network.

According to CoinShares, hashrate fell from about 1,160 EH/s in early October 2025 to 850 EH/s by early February 2026 before recovering. It describes the network’s first six-month decline since the China ban. Blockware Intelligence put the 14-day average at 946 EH/s in early October 2026, 17% below the October 2025 peak. It added that 2026 has printed the two largest downward difficulty adjustments since 2021: -11.2% in February and -10.1% in June.

This is the system working as designed. Lower difficulty makes blocks cheaper to find for the miners that remain, which helps their margins recover. A hashrate dip is therefore a sign of stress among miners, not a failure of the protocol. Blocks keep arriving at roughly the target pace.

Indicator	What it measures	What a fall usually signals
Hashrate	Total computing power on the network	Less efficient miners shutting down or redirecting power
Difficulty	How hard it is to find a valid block	Network adjusting to lower hashrate
Hashprice	Daily revenue per PH/s	Squeezed miner margins
Block reward plus fees	Revenue per block	Post-halving income, or low fee demand

Trend 3: Miners pivot to AI and high-performance computing

The most visible structural shift is that mining companies are selling their real assets, which are power contracts, cooling and data center shells, to AI customers.

CoinShares’ Q1 2026 report counted more than $70 billion in cumulative AI/HPC contracts announced across the public mining sector. It said listed miners could draw as much as 70% of revenue from AI by the end of 2026, up from roughly 30%. The report frames the move as economic: hashprice sits near cyclical lows, while AI infrastructure offers higher and steadier returns.

TheEnergyMag estimates that public miners shed about 56 EH/s of realized hashrate in the first half of 2026, a 15% contraction, with some of the freed power redirected to AI.

For readers tracking mining stocks or network security, the implication is that hashrate and miner revenue can now diverge. A miner’s share price may reflect its AI contracts as much as its Bitcoin output. Cryptela’s coverage of the overlap includes ceτi AI’s acquisition of Big Energy Investments for North American high-performance computing, and its $80M NVIDIA GPU purchase. Both are company announcements, and readers should weigh them as such.

Trend 4: Power access becomes the scarce asset

CoinShares’ Q2 2026 report points to US restrictions and grid constraints that are making already-energised sites more valuable. The logic holds for AI and mining alike: a site with a grid connection, permits and available megawatts takes years to replicate.

The Cambridge Digital Mining Industry Report (April 2025) gives useful context. It estimates Bitcoin mining uses about 138 TWh of electricity a year, with around 39.8 MtCO2e of attributable emissions. The US accounts for 75.4% of reported global mining activity. Sustainable sources make up 52.4% of the mix, and natural gas is the single largest source at 38.2%.

Building a new site also means getting power in place before the grid arrives. Guides such as temporary power for data center builds show how fuel logistics and generator planning enter the picture during construction, which is one reason energised capacity commands a premium.

The IMF working paper No. 26/146 (July 2026) adds a policy angle. It uses hardware import data from China, Hong Kong and Taiwan and finds that global factors, namely crypto prices and hardware costs, drive mining surges. Domestic factors such as electricity prices and temperature shape where mining happens. The paper also warns that cheap, often subsidized electricity can encourage mining and create financial and environmental distortions, so regulators are likely to keep examining power policy.

Trend 4: Power access becomes the scarce asset

Trend 5: Efficiency is the dividing line

Not every miner is losing money. Efficiency, measured in joules per terahash (J/TH), separates survivors from the rest.

A November 2, 2026 analysis from Shattered puts hashprice near $33 and says a 13.5 J/TH machine breaks even around $0.10/kWh before other costs, while older 18.5 J/TH equipment needs power below about $0.074/kWh. SimpleMining says sub-15 J/TH hardware at $0.07/kWh hosting can reach margins of roughly 32% to 52%, while home miners on $0.16 to $0.20/kWh residential power operate at a loss. Those are one company’s scenario figures, tied to a Bitcoin price near $78,000, which does not match the price levels CoinShares reported for mid-2026. Use them to understand the direction of the effect, not as a forecast.

The ING Mining profitability table shows the same pattern for 49 miner models on November 7, 2026. At 8¢/kWh, 12 models earned more than their electricity costs and 37 lost money. At 12¢/kWh, none covered electricity. The table excludes downtime, cooling and purchase cost, so real results are weaker.

Miner models covering electricity costs at two rates

Is crypto mining still profitable?

Yes, for some participants. Mining is still active, but profit depends on your electricity rate, your hardware and the Bitcoin price on the day you are paid.

A worked example makes it concrete. ElectroIQ illustrates that a machine drawing 3,000 watts for 24 hours uses 72 kWh. At $0.08/kWh that costs about $5.76 a day, and at $0.25/kWh it costs $18 a day. ElectroIQ rounds the first figure to about $5.70. If difficulty rises 20%, daily revenue of $10 falls to about $8.33 before any costs.

For a quick check, use the hashprice method:

Take the current hashprice in dollars per PH/s per day.
Multiply by your machine’s hash rate in PH/s. The ING table shows a 100 TH/s machine earning about $3.40 a day at $33.98 hashprice, before power.
Subtract electricity: kilowatts times 24 hours times your rate per kWh.
Account for pool fees, cooling, downtime and the purchase price spread over the machine’s life.

If step 3 already erases the result, the machine is not viable at your power rate. Home miners with residential rates usually fall into this group. Prices and difficulty move constantly, so repeat the calculation with fresh numbers. Cryptela has covered tools for this, such as Mining Now’s real-time mining insights and profit analysis platform.

Fees, halvings and the long-run revenue model

The block reward halves roughly every four years, and the original design assumes transaction fees will grow to fill the gap, as the Dynamics of Bitcoin mining paper explains. After the April 2024 halving, Binance Academy notes that activity from protocols such as Ordinals and Runes has at times pushed fee revenue near or above the subsidy on busy days. Fee income is volatile, though, so it supplements the reward and does not replace it.

Cloud mining and pools: what readers should know

Many retail readers meet mining through pools and cloud mining contracts. A pool combines the hash power of many miners and splits rewards. Cloud mining rents hash power from a provider instead of buying hardware.

If you are exploring these, the same economics apply. A contract’s return still depends on hashprice, difficulty and the provider’s fees, and there is no guarantee that revenue exceeds cost. Cryptela’s blog has published company-submitted announcements on this, including PEGA Pool and cloud mining from home. Those pieces present the companies’ own claims. Read the terms, the fee structure and the regulatory status of any provider before committing money, and only risk funds you can afford to lose.

Where DeFi fits, and how it differs from mining

Readers often search for mining and DeFi (decentralized finance) together, but they are different activities. Mining secures proof-of-work chains using hardware and electricity. DeFi uses smart contracts, programs on a blockchain, to offer lending, trading and staking without a traditional intermediary.

The headline DeFi indicator is total value locked (TVL). As MyEtherWallet explains, TVL is the dollar value of assets deposited in a protocol’s smart contracts, calculated by multiplying each asset’s amount by its market price. Because it is priced in dollars, TVL rises and falls with token prices even when no one deposits or withdraws. That is why it is a gauge of activity and trust, not a full measure of a project’s health. Pair it with audits, user counts and fee revenue.

Mining and DeFi do connect in one way: both react to crypto prices. A falling market squeezes miners’ margins and shrinks TVL at the same time.

What to watch next

A few indicators will tell you whether the squeeze eases:

Hashprice: a sustained recovery from the mid-$20s to the $30s would improve margins across the fleet.
Difficulty adjustments: a run of positive adjustments suggests miners are returning.
AI revenue share: the more miners rely on AI contracts, the less their results track Bitcoin.
Power policy: grid rules and electricity pricing in the US and elsewhere, which the IMF flags as a policy issue.
Fee revenue: whether fees consistently contribute a bigger share of block income.

Bitcoin’s price remains the dominant variable, and no one can predict it. Treat every forecast, including those cited here, as a scenario.

Reaching crypto readers with your mining news

If your company builds mining hardware, hosting, or related infrastructure and has news worth sharing, Cryptela offers a submit article option at a one-off price of 99 USD, as well as sponsored articles, press releases and banner placements through its advertising page. Sponsored and submitted pieces are labeled separately from Cryptela’s independent editorial coverage.

FAQ

Is crypto mining still profitable in 2026?

It can be, but only under the right conditions. One November 7, 2026 dataset shows that at 8¢/kWh, 12 of 49 Bitcoin miner models earned more than their electricity costs, and at 12¢/kWh none did. That table leaves out cooling, downtime and hardware cost, so home miners on residential power should expect to lose money.

How does cryptocurrency mining work?

Miners use specialized computers to compute hashes until one produces a valid result for the next block. The winner earns new coins plus transaction fees. Bitcoin adjusts difficulty about every two weeks to keep blocks arriving roughly every 10 minutes.

Why are Bitcoin miners moving into AI?

CoinShares says the shift is largely economic: hashprice sits near cyclical lows, while AI infrastructure offers higher and steadier returns. Miners already hold power contracts, cooling and data center space, which AI customers need.

What is the difference between mining and DeFi?

Mining uses hardware and electricity to secure proof-of-work blockchains. DeFi uses smart contracts to provide financial services such as lending and trading. Their indicators differ too: hashprice and difficulty for mining, total value locked for DeFi.