Explainers
What the XRP Ledger is, and what its on-chain data shows
The XRP Ledger is a public blockchain that has run since 2012 without mining or staking: each server chooses the validators it trusts, and a ledger version is final when enough of them agree. It has an exchange built into the protocol, destroys the fee every transaction pays, and requires every account to hold a minimum balance. It is also separate from Ripple, the company most associated with it, which the ledger’s documentation describes as one contributor among several.
A ledger built in 2011 and 2012
According to xrpl.org, the ledger was built over 2011 and early 2012 by Jed McCaleb, Arthur Britto and David Schwartz, and 100 billion XRP existed from the start. Its decentralised exchange has operated continuously since the launch in 2012, the same site says.
Each ledger version holds the complete current state of balances and objects, the set of transactions applied since the previous version, and metadata linking it to its parent, so a new server does not need the full history to know the present state. The consensus documentation records that ledgers 1 to 32,569, roughly the first week, were lost in a mishap, which makes ledger 32,570 the earliest available anywhere; because every version carries the whole state, the missing ledgers are not needed to know the present one.
The server software, which xrpl.org now calls xrpld, is open source. Any change to how transactions are processed, bug fixes included, is introduced as an amendment that validators vote on. An amendment that holds more than 80% support for two weeks is enabled permanently, and only another amendment can switch it off.
Consensus with a list of trusted validators
Each server operator configures a Unique Node List, the set of validators that server trusts not to collude. In each round, validators share the transactions they propose for the next ledger, move their proposals towards those of the validators they trust, and repeat until enough of them agree. xrpl.org’s introduction says a new ledger is created when 80% of validators agree, and that the process takes four to six seconds. A validated ledger is final.
The failure rules follow from that threshold. Consensus carries on while fewer than 20% of trusted validators are faulty; confirming an invalid transaction would need more than 80% of them to collude; and between those two points the network stops making progress rather than splitting. The lists themselves have to agree as well. Research cited on xrpl.org found that, in the worst case, two servers’ lists need 90% overlap to avoid a fork, against the 60% that was first thought enough.
Most servers use a default list from a publisher. The xrpl.org FAQ, read on 22 November 2026, names the XRPL Foundation and Ripple as known publishers and says any participant can choose another. It gives the network more than 150 validators, more than 35 of them on the default list and one of those run by Ripple. Validators are not paid: running one requires no fees or XRP, and the FAQ says Ripple avoids paying XRP for validation so that rewards do not distort validators’ behaviour. There is nothing to stake.
Reserves, and a fee that is destroyed
Two charges in XRP protect the shared ledger from spam. The first is a reserve. To exist, an account must hold a base reserve, 1 XRP on mainnet when read, plus an owner reserve of 0.2 XRP for each object it owns in the ledger, such as an offer, a trust line or an escrow. Reserved XRP cannot be sent to anyone else, though it can pay transaction fees, and deleting the account recovers part of it. Validators can change the reserve settings through fee voting.
The second is the transaction cost. The minimum for a standard transaction is 10 drops, a drop being one millionth of an XRP, and the cost rises with load: a server ignores transactions that pay less than its load-based threshold and queues those that pay less than the open-ledger cost, which rises exponentially once a ledger passes its soft limit on transactions. The XRP paid goes to no validator and no other party. It is destroyed.
The exchange built into the ledger
Trading on the ledger uses a native transaction type, the Offer, which is, in effect, a limit order to exchange a stated amount of one asset for a stated amount of another. Matching offers are consumed best rate first, and whatever is not filled rests in the ledger as a passive offer for later orders and cross-currency payments to take. Tokens are identified by issuer and currency code together, so two tokens with the same code from different issuers are different assets, and anyone can issue one. xrpl.org advises checking the issuer before buying, since a token may turn out to be worthless.
Since 22 March 2024, when the AMM amendment went live, the ledger has also had automated market maker pools, at most one per pair of assets, which price swaps from a formula over two reserves (how an AMM sets a price). A single transaction can fill against offers, an AMM or a mix of both, whichever is cheaper.
The ledger and Ripple, the company
xrpl.org’s account of the relationship is brief. In September 2012 McCaleb and Britto, with Chris Larsen, formed the company now called Ripple, then named OpenCoin Inc., and gave it 80 billion XRP in exchange for developing the ledger. The company has since sold XRP regularly and in 2017 placed 55 billion XRP in escrow. The ledger itself was first called Ripple and the company registered as Ripple Labs; to avoid confusion between the names, the asset came to be called XRP in every context, and the company later shortened its own name to Ripple.
On control, the FAQ says the ledger is a public blockchain and not a private one owned by Ripple, that Ripple is a contributor whose rights are the same as other contributors’, and that changes affecting transaction processing or consensus need the approval of at least 80% of the network. A new XRPL Foundation, incorporated in France, lists Ripple among its founding members alongside XRPL Commons, XRPL Labs and XAO DAO, and was set to take over the default validator list that the previous foundation published. The XRP trademark, registered in 2013 with OpenCoin and Ripple Labs as assignees, has been assigned since 2022 to the XRPL Foundation, xrpl.org says.
The SEC case, as the record shows it
On 22 December 2020 the SEC sued Ripple Labs and two executives, Christian Larsen and Bradley Garlinghouse, alleging that they had raised over $1.3 billion through an unregistered securities offering of XRP that began in 2013.
On 13 July 2023 Judge Analisa Torres of the Southern District of New York ruled on cross-motions for summary judgment (opinion and order). The court granted the SEC’s motion on Ripple’s institutional sales, made under written contracts to counterparties such as institutional buyers and hedge funds, and the defendants’ motion on programmatic sales through exchanges, on other distributions such as payments for services, and on Larsen’s and Garlinghouse’s own sales. Its analysis turned on the transactions rather than the token: it held that XRP, as a digital token, is not in itself an investment contract, and examined each kind of sale on its own circumstances. Programmatic buyers, it found, had bought in blind bid/ask trades and could not have known whether their money went to Ripple at all.
The final judgment of 7 August 2024 imposed a civil penalty of $125,035,150 on Ripple and an injunction against further violations of the Securities Act’s registration provisions, and both sides appealed. On 8 May 2025 the SEC and Ripple filed a settlement agreement with the district court and asked it to indicate whether it would dissolve the injunction and release the escrowed penalty, $50M to the SEC and the rest to Ripple; the SEC said its decision rested on its efforts to reform its approach to the industry, not on the merits. The court declined, on 15 May and, on a renewed request, on 26 June 2025. On 7 August 2025 the SEC announced a joint stipulation dismissing both appeals, which resolved the action, and said the final judgment, with its penalty and injunction, would remain in effect.
Every claim in the case concerned offers and sales of XRP by Ripple and its executives.
Documented incidents
A halt in late 2024 was short. A vulnerability report on xrpl.org says that on 25 November 2024 several nodes across the network crashed and restarted at about the same time from 13:39 UTC, and the network processed no transactions for about ten minutes, resuming at 13:49. A January 2024 change to the server’s caching code had moved a type check, so that in certain transactions an object of an unexpected type crashed the server instead of being treated as missing. Versions 2.0.1 to 2.2.x were affected, no funds were lost, and the report notes that the consensus model favours safety over progress while nodes are unstable.
The second incident did not touch the ledger at all. On 21 April 2025, several versions of xrpl.js, the JavaScript library for the ledger, were published to npm carrying code that sent private-key material to an attacker’s website, after a maintainer of the package was phished and the attacker used the compromised npm account. According to the disclosure report, the ledger and its codebase were not affected, no GitHub repository was compromised, and the malicious versions were deprecated on 22 April, the day the attack was discovered. The weak point was a library that applications use to create and handle keys, a pattern how crypto gets hacked follows through the wider record.
What the on-chain data shows
As of 22 November 2026, per DefiLlama, the XRP Ledger, listed as XRPL, had $29.9M of total value locked, 23.8% lower than thirty days earlier, with 18 protocols listed and $844.4M of stablecoin supply on the chain (data page). The total is the value of assets deposited in DeFi protocols on the chain at the snapshot’s prices, leaving out liquid staking, bridges and deposits DefiLlama flags as double-counted (what TVL measures). Stablecoin supply is about 28 times that total (computed), so at most a small part of the stablecoins circulating on the ledger can be sitting in deposits the total counts; where stablecoin supply lives covers how that supply figure is built.
The list of largest protocols is on a different basis, and on this chain it runs far above the total. Maple, filed under Lending, stands at $238.0M on the ledger, about eight times the chain figure; the chain total therefore cannot include all of it, and the snapshot does not record why. Three of the next entries are bridges, a category the chain total leaves out: Coinbase Bridge at $147.8M, which DefiLlama describes as wrapped tokens backed one-for-one by assets Coinbase holds; FAssets at $134.9M, described as a bridge on Flare for minting wrapped versions of assets such as XRP; and Wrapped at $16.7M. Doppler Finance, at $85.8M, is filed under CeDeFi.
The entry closest to the ledger’s own design is XRPL DEX at $26.7M, filed under Dexs, which on its own equals 89.0% of the chain figure (computed). DefiLlama describes it as the ledger’s AMM pools. Its description does not mention the offers resting on the ledger’s order book, and the snapshot does not say whether they are counted.
The site has no hack page for the XRP Ledger, because DefiLlama’s record does not yet hold the twenty incidents a chain page needs.