Litecoin (LTC): a complete guide to the coin and its mining

Litecoin is one of the oldest proof-of-work networks still running. The client was released on 7 October 2011 by Charlie Lee, at the time an engineer at Google and later director of engineering at Coinbase. The code came from Bitcoin, but the algorithm was swapped for Scrypt and the block time was cut four times over, down to 2.5 minutes.

The idea was practical: a network for fast, cheap payments where a confirmation arrives within minutes and the fee is measured in fractions of a cent. Over fourteen years Litecoin also became a testing ground for protocol work: SegWit was activated here before Bitcoin got it, and the confidential extension blocks known as MWEB went live in 2022, well ahead of comparable attempts on other large networks.

For a hardware owner the picture in 2026 is unusual. The Litecoin network and the Dogecoin network are secured by the same work through merged mining, and at current prices it is Dogecoin that brings in roughly four fifths of a Scrypt miner revenue. That inverts the usual logic: planning income from the LTC price alone means being wrong by a wide margin.

This page is built as a reference: network parameters, the history of protocol decisions, a ranking of Scrypt ASICs from our catalogue, pools, wallets, mining economics, a breakdown of the MWEB incident in the spring of 2026 and a separate list of figures that the big aggregators still publish in an outdated form.

• 2011Charlie Lee publishes the client on 7 October. It is based on Bitcoin code, the algorithm is replaced by Scrypt taken from the Tenebrix project, the block time is cut to 2.5 minutes and the supply cap is raised to 84 million coins.
• 2013The November rally: the price doubles in a day and the coin reaches the top of the market for the first time. Mining still runs on graphics cards.
• 2014The first Scrypt ASICs appear and graphics cards leave the network. In September Dogecoin switches to AuxPoW and starts being mined together with Litecoin, an initiative that came out of the LTC community.
• 2015First halving on 25 August: the reward drops from 50 to 25 LTC.
• 2017SegWit is activated on 10 May, three months before Bitcoin. The signature moves out of the main part of the transaction. The same month brings the first Lightning payment on the Litecoin network.
• 2017In December Charlie Lee sells and donates his entire LTC holding, explaining it as a conflict of interest between developing the project and holding the coin.
• 2019Second halving on 5 August: 25 LTC becomes 12.5 LTC.
• 2021May brings the all-time price high of around 410 dollars. In September a fake press release about a partnership with a large retail chain pushes the quote up by roughly 30 percent within minutes, followed by a sharp retracement.
• 2022MWEB is activated on 19 May: extension blocks with hidden amounts and addresses. In June the five largest exchanges in South Korea delist LTC, classifying the coin as privacy enhanced.
• 2023Third halving on 2 August: the reward falls from 12.5 to 6.25 LTC. The LTC part of a Scrypt miner revenue is cut in half from one block to the next.
• 2025The year of the institutional turn. On 18 July a public treasury company starts up holding more than 929,000 LTC. On 28 October the first spot exchange traded fund on Litecoin begins trading on Nasdaq. On 9 November the network passes block 3,000,000. On 6 December the all-time hashrate high is recorded at around 3.94 PH/s.
• 2026In the spring the network goes through the most serious technical incident in its history. On 19 March an exploit of an MWEB vulnerability is found and confirmed, with the attacker creating a fictitious withdrawal of 85,034.47 LTC. On 25 April a second attempt causes a chain reorganisation of 13 blocks. The full fix ships on 7 May in Litecoin Core 0.21.5.5.
• 2026In July the settlement depository of the Deutsche Boerse group adds LTC to institutional custody. At the same time inflows into the spot fund stay below 10 million dollars over its first eight months, so access through a regulated wrapper has not created demand by itself.

The accounting model is the same as in Bitcoin: the network stores no account balances, only a set of unspent outputs. A transaction points at earlier outputs, spends them and creates new ones, and the right to spend is proven by a signature. The sum of all unspent outputs is the coin supply.

A block is limited to 4,000,000 weight units, with the base part up to one megabyte. Since blocks are found four times more often than in Bitcoin, daily capacity is roughly four times higher at the same block size. The network is far from full: the transfer fee sits at fractions of a cent and there is almost always room in a block.

Alongside the main chain runs MWEB, an extension block. Coins enter it with a peg-in operation and leave with a peg-out, and inside it amounts and addresses are hidden. It is neither a separate chain nor a second layer: extension blocks are secured by the same miner work, and older nodes see them as valid but opaque data.

Every node checks a found block independently: signatures, the absence of double spends, the size of the reward and the fact that the hash is below the current target. The incident in the spring of 2026 showed the cost of a gap in that check: a logic hole in the validation of an MWEB withdrawal briefly allowed coins to be created out of thin air, and closing it took emergency releases and coordination with the pools.

1. The wallet signs a transactionThe key owner builds a transfer and signs it. Only the signature and public data go out to the network, the private key is never published.
2. The transaction spreads across nodesEach node checks it against its own rules and puts it in its own waiting pool. With the network lightly loaded, that queue is almost always empty.
3. The pool assembles two templates at onceThe mining pool prepares a Litecoin block and at the same time embeds a reference to a Dogecoin block in it. That is the basis of merged mining.
4. The ASIC grinds through Scrypt hashesThe hardware changes service fields in the header and computes Scrypt until the result falls below the target. The same work counts for both networks.
5. The share credits whichever chain target it beatIf the hash clears the Litecoin target, an LTC block is found. If it only clears the easier Dogecoin target, a DOGE block is found. Often both happen at once.
6. Nodes accept the blockEach node validates the block on its own. The first transaction creates the reward: the 6.25 LTC subsidy plus the fees of the transfers included.
7. Difficulty is recalculatedEvery 2,016 blocks the network compares the actual elapsed time with the expected 3.5 days and adjusts the target so that a block is again found in 150 seconds.

Scrypt is a key derivation function proposed by Colin Percival in 2009 to protect passwords. Unlike SHA-256 it deliberately demands memory: to compute a result you have to build and hold a large array of pseudorandom data in working memory. Litecoin uses it with parameters that put roughly 128 kilobytes behind a single computation.

The intent was to put an ordinary computer and a purpose-built chip on the same footing: memory is expensive in silicon, so an ASIC advantage would stay small. In practice the intent failed. Scrypt ASICs arrived as early as 2014, and 128 kilobytes turned out to be far too little to act as an obstacle. Today the gap against a graphics card is measured in thousands of times.

Mining Litecoin on a graphics card or a processor makes no sense in 2026 at any electricity rate. A modern machine puts out tens of gigahashes at an efficiency near 0.15 joules per megahash, while a graphics card works at the level of single megahashes. Any service promising otherwise is selling either a cloud contract or arithmetic built on numbers that do not exist.

The main practical difference between Scrypt and SHA-256 is not the function itself but the economics around it. One stream of work serves several networks at once: Litecoin, Dogecoin and a set of smaller chains such as Bellscoin, Junkcoin and Luckycoin. That costs no extra energy, so income is counted as the sum across every chain credited, not from a single coin.

Strong points: real competition between vendors, three independent makers in the top segment and extra income from auxiliary chains. Weak points: the model line-up is an order of magnitude smaller than on SHA-256, the second-hand market is thinner, and income depends on the price of two coins at once, one of which is a meme coin and unpredictable.

Full description of the Scrypt algorithm →

The supply cap is 84,000,000 LTC, exactly four times that of Bitcoin, because blocks are found four times as often. At the end of July 2026 around 77.4 million coins were in circulation, roughly 92 percent of the cap. The arithmetic is simple and easy to check: 840,000 blocks at 50 LTC, the same number at 25, the same at 12.5 and the current stretch at 6.25.

The reward halves every 840,000 blocks. The steps already passed: 50 LTC from 2011, 25 LTC from 25 August 2015, 12.5 LTC from 5 August 2019 and 6.25 LTC from 2 August 2023. The next halving is tied to block 3,360,000. As of late July 2026 some 208,000 blocks were left to it, with an expected date in late July 2027. Nobody knows the exact day: it depends on how fast blocks are actually found, and right now a block takes 2 minutes 26 seconds, faster than the target of 150 seconds.

Annual issuance at the current reward is roughly 1.31 million coins, so supply inflation sits near 1.7 percent a year. After the 2027 halving that figure drops to about 0.84 percent. The protocol has no coin burning and no other supply reduction mechanism.

Fees play almost no part in block economics. The average transfer fee stays around 0.00005 LTC, that is fractions of a cent, and against a subsidy of 6.25 LTC its contribution is negligible. That is convenient for users and inconvenient for miners: income from the Litecoin network is effectively the subsidy alone.

The point that changes the whole calculation: the larger half of the income comes from Dogecoin. The Litecoin network issues around 3,600 LTC a day, the Dogecoin network around 14,400,000 DOGE. At late July 2026 prices, on the order of 45 dollars per LTC and 0.07 dollars per DOGE, that is about 162 thousand dollars against about 1 million. The Litecoin share of daily revenue across the whole Scrypt industry works out near 14 percent, with Dogecoin supplying the rest.

The full table of Scrypt models from our catalogue, sorted by efficiency: the lower the joules per megahash, the longer a machine stays profitable as difficulty grows and after the 2027 halving.

Browse miners in the catalogue →

The ranking is built on real efficiency figures from the catalogue, not on how popular a model is. The break-even thresholds are calculated from total network revenue on the order of 0.45 to 0.50 dollars per gigahash a day, taken at the end of July 2026, and they include the DOGE payout.

Third-party firmware on Scrypt is a much narrower story than on SHA-256. In our catalogue support is marked only on Bitmain machines, namely the Antminer L7 and L9. For the ElphaPex, SealMiner and Goldshell lines the catalogue records no third-party builds: those vendors entered the market recently and there are noticeably fewer open builds for their boards.

Bitmain factory firmware lets you start a machine and pick one of the preset modes, but it does not give you control of voltage and frequency on individual hash boards. That makes it impossible to bring a specific unit to its own optimum: chips differ from one another even within a single batch.

Third-party builds exist precisely for that job. Through HashCore Toolkit a supported Antminer receives firmware with manual tuning of frequencies and voltages per chain, automatic mode selection and telemetry for every board. The developer fee is 2.8 percent. In the catalogue this support is marked on the L7 and L9, as well as on the S19 and S21 series on another algorithm.

It helps to understand the limits. The AsicBoost mode belongs to SHA-256 and does not apply to Scrypt: it exploits a property of the header structure under double hashing that Scrypt simply does not have. Everything reflashing gives you on Scrypt is undervolting, precise frequency selection and stability, not a new algorithmic advantage. Promises of multiplying hashrate on Scrypt should be read as advertising.

The economic sense is direct: at a fixed tariff, cutting the energy spent per megahash pushes back the point where a machine goes into the red. For the L9 that is the difference between breaking even and a small profit on an industrial tariff. The limits are just as direct: reflashing voids the warranty, and a wrong voltage degrades the boards. Work with a backup of the factory build and raise the tuning gradually, watching temperature and the share of rejected shares.

You can only start mining Litecoin with a Scrypt ASIC. The minimum kit is the machine itself, its stock power supply, a dedicated power line, wired internet and a pool account. There is no need to run a node, the block template comes from the pool. There is no separate setup for Dogecoin either: the pool attaches the auxiliary chain itself.

Electrical supply is the first constraint. Industrial Scrypt machines draw from 3.1 to 7.8 kW, so they need a dedicated line and a breaker rated at 16 to 32 amps. A household socket will not carry that load. The DL1 Hydro at 7,823 W also needs a ready loop with a heat exchanger.

The second constraint is noise and heat. Air-cooled industrial models run at around 75 decibels and above around the clock, comparable to a vacuum cleaner that never stops. You cannot put such a machine in a living space. The catalogue has one exception: the ElphaPex DG Home 1 at 620 W and the Goldshell Mini-DOGE line, which are designed for a room.

Home mining of Litecoin in 2026 makes sense in two scenarios: learning, and reusing heat in winter when the machine partly replaces heating. As a source of income on a household tariff in Spain it does not work: the break-even point even for the best machines in the catalogue lies around 0.13 to 0.14 dollars per kilowatt hour, while home machines only work below 0.065.

One practical setup tip: connect to a pool that pays out DOGE, and check that both payouts show up in your account. If the pool pays LTC only, you lose the larger part of the revenue while saving nothing on electricity. A backup pool in the machine configuration is a must: a few hours of downtime costs more than any difference in fees.

Revenue from a Scrypt machine has three parts: the Litecoin subsidy, the Dogecoin payout from merged mining and small change from auxiliary chains. At the end of July 2026 that came to roughly 0.45 to 0.50 dollars per gigahash a day in total. The order of magnitude is easy to verify: the networks issue around 3,600 LTC and 14,400,000 DOGE a day at a combined hashrate of around 2.46 PH/s.

The split between those parts is unusual today. At prices near 45 dollars per LTC and 0.07 dollars per DOGE, Litecoin accounts for roughly 14 percent of revenue and Dogecoin for roughly 86. Per coin mined that works out at about 4,000 DOGE for every 1 LTC. The practical conclusion follows: income is more sensitive to the price of the meme coin than to the price of Litecoin itself, and that is the main risk in the model.

Network fees can be left out of the calculation. The average transfer fee is around 0.00005 LTC, and across 590 blocks a day their contribution to the reward is measured in hundredths of a percent. Plan strictly from the subsidy and the DOGE payout.

Two risk factors over the coming year. The first is difficulty: it retargets every 3.5 days and during a market rally it eats income faster than the price rises. The second is the halving in late July 2027. It will cut in half the part of revenue that comes from Litecoin. At current price ratios that is a loss of around seven percent of total income rather than half of it, but if DOGE falls against LTC the blow grows in proportion. A machine bought in the second half of 2026 is best assessed on a payback horizon no longer than a year, so that the exit point does not sit beyond the halving.

Open the profitability calculator →

For a miner the rule is simple: the address the pool sends payouts to must be fully yours. An exchange address is convenient, but if the account is frozen your mined coins go with it. A hardware wallet plus a seed phrase written down offline closes that risk for sensible money. Remember that payouts arrive in two coins: the DOGE address has to be yours as well.

On address formats: the modern ltc1 takes less block weight than the old L... style, so spending from it is cheaper. MWEB addresses in the ltcmweb1 form are not supported everywhere, and hardware wallets do not handle them at all. After the incident in the spring of 2026, withdrawals from MWEB require six confirmations, a rule built into the client from version 0.21.5.5 onward.

Litecoin trades on every major venue and remains one of the most liquid coins outside the top twenty. As of 31 July 2026 daily trading volume was around 260 to 280 million dollars against a capitalisation near 3.5 billion, so roughly eight percent of the market turns over in a day. Among venues available in the EU, Kraken, Bitvavo, Bit2Me and Coinbase are the common choices, and Binance, OKX and Bybit among the global ones.

Regulated access arrived recently and works less well than expected. The spot exchange traded fund on LTC launched on Nasdaq on 28 October 2025, but over its first eight months it gathered less than 10 million dollars of inflows, an order of magnitude below comparable Bitcoin and Ethereum products in their first quarter. In July 2026 the settlement depository of the Deutsche Boerse group added LTC to institutional custody.

Keep regional restrictions in mind separately. Since June 2022 the largest South Korean exchanges have not traded LTC because of MWEB, and as of late July 2026 that delisting still stands. It does not affect mining, but it does affect where you can sell and how deep the order book is during the Asian session.

We sell hardware, not the coin. This page is a reference: pick a venue yourself based on its licence in your jurisdiction and its fiat withdrawal terms. Check the withdrawal address character by character, especially when you handle two coins at once: sending DOGE to an LTC address cannot be undone.

The main use case is a fast, cheap transfer. Confirmation arrives within minutes, the fee stays around one five-hundredth of a cent and there is almost always room in a block. Payment processor reports put Litecoin consistently among the top three coins by transaction count, and in some months of 2025 it accounted for 20 to 30 percent of payments excluding stablecoins.

The second use case is optional privacy. MWEB lets you hide amounts and addresses while staying on the same chain. The scale is still small: at the end of July 2026 the extension blocks held around 473 thousand LTC, less than one percent of supply, with roughly 313 operations a day. Using it is voluntary and ordinary transfers stay fully transparent.

The third use case is as a technical proving ground. Litecoin has historically adopted upgrades ahead of Bitcoin: SegWit was activated here on 10 May 2017, three months before the main network, and the first Lightning payment also went through in May 2017. The other side of that role is visible in MWEB: the network is first to catch both other people mistakes and its own.

A fourth use case appeared in 2025, the coin as a reserve asset. A public treasury company holding more than 929,000 LTC started work on 18 July 2025, and a spot exchange traded fund has traded on Nasdaq since 28 October of the same year. For a miner this matters indirectly, through demand: the more reasons there are to hold the coin, the steadier the price his revenue is measured in.

Network security rests on the cost of an attack, and here Litecoin has a structural quirk. A hashrate of around 2.46 PH/s is smaller than Bitcoin on its own, but it does not protect one chain: the same work also holds up Dogecoin. A 51 percent attack does not let anyone steal other people coins or issue extra ones, at most it reverses the attacker own recent transactions, which is exactly why exchanges wait for confirmations.

The weak spot is not the hardware but how it is organised. Three pools find around 78 percent of blocks over a weekly window, above the concentration level in Bitcoin. Each individual machine still belongs to its owner and switching to another pool takes minutes, which is what limits the practical risk. The number of independent nodes is small: crawlers show on the order of 2,000 reachable, an order of magnitude fewer than on the Bitcoin network.

In the spring of 2026 the network went through the most serious incident in its history, and it is worth reading from the official developer report. On 19 March an exploit of a flaw in MWEB withdrawal validation was confirmed at block 3,073,882: the attacker produced a fictitious peg-out of 85,034.47 LTC. Developers working with the pools blocked the attempt to spend those coins, ran a recovery transaction on 24 March and finished rebalancing on 25 March. The finder was awarded a bounty of 850 LTC.

On 25 April 2026 a second attempt followed. It used a different flaw, a denial of service on modified blocks, which let an invalid branch grow to 13 blocks, from 3,095,930 to 3,095,943, rolling back roughly 32 minutes of transactions. The valid chain was restored through miner coordination and version 0.21.5.4 shipped the same day. The cause was named as a logic error in the MWEB protocol itself, not a capture of hashpower. The full fix went into version 0.21.5.5 of 7 May 2026. Third parties were hit: a cross-chain service lost around 11,000 LTC in equivalent value and suspended operations with the coin. The minimum safe node version today is 0.21.5.4, the recommended one 0.21.5.5.

The block reward is quoted as 12.5 LTC. That value applied until 2 August 2023. From block 2,520,000 the subsidy is 6.25 LTC. The error shows up even in calculators and overstates the Litecoin part of calculated income exactly twofold.

The halving is dated to August 2027. The halving is tied to block 3,360,000, not to the calendar. The actual block time at the end of July 2026 is around 2 minutes 26 seconds instead of the 150 second target, so the estimated date moved to late July 2027. Some 208,000 blocks were left.

The 2027 halving is presented as cutting miner income in half. Only the part of revenue that comes from Litecoin will halve. At late July 2026 prices that is around 14 percent of the income of a Scrypt machine, with Dogecoin supplying the rest, and Dogecoin has a fixed reward and no halvings. If prices hold, the total blow is around seven percent, but it grows if DOGE falls against LTC.

The merged mining ratio is given as 800 DOGE per 1 LTC. That number comes from the era of a high Litecoin reward. At the current 6.25 LTC per block and 10,000 DOGE per block the arithmetic runs like this: 576 LTC blocks a day against 1,440 DOGE blocks, so roughly 4,000 DOGE for every LTC mined.

The Antminer L7 is called the current Scrypt flagship. A 2021 machine at 0.361 J/MH trails the catalogue leaders by 2.4 times and loses money on European tariffs. The top rows in 2026 belong to the SealMiner DL1 and the ElphaPex DG2+, and Bitmain has already released the next L11 generation.

The Antminer L9 is presented as a new release. The model came out in May 2024. At 0.10 dollars per kilowatt hour it runs near zero, and on household European tariffs at a loss. Selling it as a new release is not honest.

The network hashrate is rounded to 1 PH/s or 2 PH/s. Those are outdated values. The all-time high of around 3.94 PH/s was recorded on 6 December 2025, and on 31 July 2026 the network ran in the 2.46 to 2.80 PH/s range depending on the source. The trend since December has been downward.

Circulating supply is taken as around 78.8 million LTC. Some aggregators publish that figure, but it does not match the issuance formula. From a block height of around 3,151,800 the calculation gives roughly 77.4 million LTC, and both the specialised halving counters and the large aggregators agree with that.

MWEB is described as mandatory privacy across the whole network. MWEB is optional. At the end of July 2026 the extension blocks held around 473 thousand LTC, less than one percent of supply, with roughly 313 operations a day. Ordinary Litecoin transactions are fully transparent.

Sources write that the network has run for 14 years without a single incident. The claim of uninterrupted operation was true until the end of 2025. In the spring of 2026 the network went through an exploited MWEB vulnerability with a fictitious withdrawal of 85,034.47 LTC on 19 March and a chain reorganisation of 13 blocks on 25 April. A reference without that section is out of date.

Sources claim merged mining needs separate hardware or reduces LTC income. Not true. AuxPoW credits the same work to two chains, no extra energy and no second machine are needed, and Litecoin income is not reduced by it. Merged mining is switched on at the pool.

Sources claim Korean exchanges have relisted LTC. There is no confirmation of that. The delisting by the five largest South Korean venues, carried out in June 2022 because of MWEB, was still in force at the end of July 2026. Campaigns for a return continue, but no decision has been taken.