Lightning Network Grows Up: Custody, Credit, and an IPO

Eighteen months ago, running a Lightning node meant a Raspberry Pi in a closet, a laptop open at 2 a.m. chasing inbound liquidity, and a community that measured success in sats routed for friends. In July 2026, the same network is backing a revolving credit line, a yield product marketed to corporate treasuries, and the balance sheet of a company that trades on the New York Stock Exchange. Bitcoin itself has been trading in the mid 60,000 dollar range this month, down sharply from its 2026 high, yet the infrastructure being built on top of its payment layer has never looked more like traditional finance.

This is not another Lightning growth-metrics story. Nodes and public capacity still matter, and we will get to the numbers, but the more interesting shift in mid-2026 is who is showing up to build on the network and what they are building. Custodians, hardware wallet makers, licensed banks, and a freshly public trust company are wrapping Lightning’s instant settlement in the products enterprises already understand: custody accounts, credit lines, yield programs, and point of sale terminals. The result is a network that increasingly looks less like a hobbyist project and more like plumbing for the next generation of payment rails, for better and for worse.

What follows traces that stack from the bottom up: who got regulatory approval to build it, what they actually built, how much of it is genuinely new versus repackaged infrastructure from 2023 and 2024, and what it costs users in the one currency Bitcoin was supposed to make unnecessary, namely trust in a third party.

From Hobbyist Node to Balance Sheet Product

Lightning was designed to let two parties move bitcoin back and forth off-chain, opening a payment channel with an on-chain transaction and settling the net result to the base layer whenever they close it. That design never required a bank, a custodian, or a regulator; anyone with a laptop and some patience could run a node. For years, that is mostly who did.

The network’s public numbers still tell an ambivalent story. As of the most recent published reading, Lightning’s public capacity sat at roughly 4,898 BTC spread across 41,080 channels and 17,438 nodes, according to Spark’s 2026 state of the network research, a node count that has fallen by close to a sixth since its 2022 peak even as capacity has trended higher. Private and unannounced channels, the kind used by mobile wallets, exchanges, and liquidity providers, are estimated to hold at least twice that much again. Growth, in other words, is not coming from more hobbyists opening more channels. It is coming from fewer, larger participants moving real capital.

That distinction matters because the identity of those larger participants changed dramatically over the past eight months. Where 2025’s headline was Binance and OKX quietly parking bitcoin into Lightning channels, 2026’s headline is a wave of named, regulated, sometimes publicly traded companies building financial products directly on top of the protocol.

It is a sharp turn from Lightning’s last moment in the spotlight. The 2021 and 2022 wave of attention centered on retail tipping, social media integrations, and El Salvador’s national rollout, all genuinely bottom up, and all eventually running into the same wall: ordinary users found channel management, inbound liquidity, and node uptime too fiddly to bother with, a large part of why node counts have declined most years since. The 2026 wave is closer to the opposite bet. Instead of asking millions of individuals to learn Lightning, it asks a few dozen regulated companies to absorb that complexity once and resell the result as a simple API or a bank feature.

A Custodian Goes Public: BitGo’s Road to Lightning

No company captures that shift better than BitGo. The firm has spent more than a decade as a back-end custody and wallet-infrastructure provider for exchanges, asset managers, and corporate treasuries, the kind of company most retail users never interact with directly but that touches a meaningful share of institutional bitcoin in circulation. It priced its initial public offering at 18 dollars a share on January 22, 2026, debuting on the New York Stock Exchange under the ticker BTGO at a valuation of roughly 2.1 billion dollars, one of the first major crypto listings of the year, according to Fast Company’s coverage of the debut. Shares popped by roughly a quarter intraday before giving back most of the gain in the weeks that followed, an ordinary post-IPO pattern that nonetheless made BitGo the most closely watched crypto infrastructure stock of the first quarter.

The IPO followed a separate, arguably more consequential milestone. On December 12, 2025, the Office of the Comptroller of the Currency conditionally approved five national trust bank charter applications, including a conversion for BitGo Bank and Trust, National Association, alongside Ripple National Trust Bank, a Circle-owned de novo bank, Fidelity Digital Assets, and Paxos, per the OCC’s own announcement. Comptroller Jonathan V. Gould framed the approvals as a vote of confidence in the sector, saying that “new entrants into the federal banking sector are good for consumers, the banking industry and the economy.” BitGo’s charter conversion gave the company direct federal oversight rather than a state-by-state patchwork of money transmitter licenses.

A federally chartered, publicly traded custodian is a very different kind of company from the node operators Lightning was built for, and that is precisely the point. BitGo spent the following months turning that regulatory and capital markets standing into a suite of Lightning products, starting with custody and expanding from there.

Crypto as a Service: BitGo and Voltage Wire Lightning Into the Back Office

BitGo’s first Lightning move actually predated the IPO. By December 17, 2025, the company had added Lightning access directly to its qualified custody platform through a partnership with Voltage, a Lightning infrastructure provider, letting institutional clients create wallets, send payments, and generate invoices through simple APIs rather than running their own nodes, as Bitcoin Magazine reported at the time. BitGo CEO Mike Belshe described the pitch in blunt terms: “by offering institutional access to Lightning directly from custody, we are allowing our clients to focus on innovation instead of infrastructure.”

That custody integration became the foundation for something bigger. On May 20, 2026, BitGo folded Lightning support into its full Crypto-as-a-Service platform, the product it sells to fintechs, exchanges, and payment apps that want to offer bitcoin transactions without becoming infrastructure operators themselves, according to CryptoBriefing’s writeup. Under the arrangement, Voltage still handles node management, channel operations, and liquidity provisioning; BitGo layers on custody, compliance, and a single API that already touches more than 1,400 digital assets across over 40 blockchains. BitGo pitches the integration for a specific set of enterprise use cases:

• Bitcoin deposits and withdrawals for exchanges and wallets
• Merchant settlement at the point of sale
• Micropayments for content, data, or API access
• Loyalty and rewards programs denominated in bitcoin
• In-app bitcoin transfers between customers

For a payments startup or a mid-sized exchange, the pitch is that Lightning becomes just another rail in an existing integration rather than a separate engineering project requiring its own uptime guarantees and liquidity management headcount. It is worth being clear eyed about what this buys and what it costs, though: speed and cost improve markedly versus routing a transaction on Bitcoin’s base layer or through a card network, but the client is trusting BitGo and Voltage with operational control of the channels, a meaningfully different risk profile than self-hosting a node. That tradeoff recurs throughout this wave of institutional products.

Lightning Earn: Turning Idle Bitcoin Into a Yield Product

The clearest example of that stacking is Lightning Earn, which BitGo launched on June 11, 2026. The product lets corporate bitcoin treasuries and institutional allocators deploy their holdings as Lightning liquidity and collect bitcoin-denominated fees in return, using infrastructure from Amboss, according to the company’s own announcement. There are two ways to earn: payment routing, where deployed bitcoin sits in channels and captures a cut of forwarded payments, and liquidity leasing, where capacity is rented out to other network participants who need it. Minimum allocation is 1 BTC, and BitGo says it seeded the program with 10 BTC of its own treasury.

The infrastructure underneath is Amboss’s Rails product, which handles the operational grind that has historically kept casual node runners from earning meaningful routing income: peer selection, fee setting, and channel sizing, optimized on an ongoing basis rather than configured once and forgotten. That automation is the real product here. Lightning routing fees have always existed; what has been missing is a way for someone who is not a protocol specialist to capture them reliably. For context on the scale gap, a well-run personal node with a modest amount of capacity might earn a few dollars a month in routing fees; Amboss’s pitch is that automated, professionally managed liquidity captures meaningfully more of the available fee volume, though neither BitGo nor Amboss has published audited yield figures that investors could benchmark against traditional fixed income.

The framing echoes, deliberately or not, the pitch behind proof of stake yield: put an idle asset to work securing or servicing a network, collect a fee, let the operator handle the mechanics. Our explainer on validator economics covers the same tradeoff on the proof of stake side, where stakers weigh yield against slashing risk and operator trust; Lightning liquidity providers weigh yield against channel management risk and, in Lightning Earn’s case, custodian trust, since BitGo holds the underlying bitcoin throughout.

Amboss RailsX and the Self-Custodial Exchange Angle

Amboss is not only powering BitGo’s yield product; it is also building its own trading venue on top of Lightning. The company unveiled RailsX on January 30, 2026, at the PlanB Forum in El Salvador, describing it as the first peer-to-peer exchange built natively on the Lightning Network, per The Block’s coverage of the launch. RailsX lets users swap bitcoin for stablecoins directly, using circular self-payments across Lightning channels to execute atomic swaps without handing custody to a centralized exchange. It leans on Amboss’s existing Magma liquidity marketplace and on Taproot Assets, the Lightning Labs protocol that lets tokens like USDT ride Lightning channels alongside bitcoin.

Amboss CEO Jesse Shrader pitched the product in explicitly regulatory terms, framing it as aligned with the direction of the CLARITY Act, the market structure bill reshaping how the SEC and CFTC divide jurisdiction over digital assets in Washington. The ambition is large: global foreign exchange markets move something in the neighborhood of 9.5 trillion dollars a day, and RailsX’s pitch is that a meaningful slice of currency and stablecoin conversion could eventually settle over Lightning rather than through bank rails or centralized exchange order books.

Whether that ambition is realistic is a separate question from whether the mechanics work. What RailsX represents for this story is a second model for institutional Lightning products, one that keeps users in self-custody throughout rather than routing bitcoin through a custodian’s balance sheet, standing in contrast to BitGo’s approach above.

Voltage Credit: Borrowing Dollars Against Bitcoin Cash Flow

Voltage, the infrastructure provider doing the heavy lifting behind BitGo’s custody and Crypto-as-a-Service products, also launched a lending product of its own. Voltage Credit, announced February 19, 2026, is a revolving credit facility that lets enterprises transact over Lightning while repaying in US dollars, so a business can accept or move bitcoin without ever carrying it on its own balance sheet, according to The Block’s report. There are no origination fees; pricing is a fixed annual percentage rate on outstanding balances, and credit limits are set through revenue-based underwriting tied to a client’s transaction volume rather than a traditional credit check.

Voltage has spent several years positioning itself as unglamorous plumbing for other companies, running nodes and managing liquidity behind the scenes rather than serving consumers directly, which is exactly the kind of role that makes a credit product a logical extension: few companies have a clearer real-time view of a client’s actual Lightning cash flow than the firm routing it. Voltage CEO Graham Krizek explained the gap the product is meant to close: “until now, using Bitcoin for payments meant managing cryptocurrency on your balance sheet. Voltage Credit eliminates that tradeoff.” The underlying problem is real. Traditional banks generally will not lend against bitcoin-denominated revenue, and existing crypto-collateralized loans typically require posting bitcoin itself as collateral, which creates tax complications and price exposure that a payments business has no interest in taking on.

The product arrived a few weeks after the kind of transaction that made the case for it. On January 28, 2026, Secure Digital Markets routed 1 million dollars to Kraken over Lightning using Voltage’s infrastructure, settling in roughly 0.43 seconds, at the time the largest publicly reported single Lightning payment. That is the scale institutional Lightning traffic is starting to operate at, and Voltage Credit is a bet that enterprises moving that kind of volume would rather borrow dollars than manage bitcoin exposure to do it.

Cold Storage Catches Up: Blockstream Jade and Lightning

Every product so far in this piece routes bitcoin through somebody else’s infrastructure, whether that is BitGo’s custody, Voltage’s nodes, or Amboss’s liquidity engine. Blockstream took the opposite approach. On March 10, 2026, the company shipped version 5.2.0 of its Blockstream app, making the Jade hardware wallet the first cold storage device able to send and receive Lightning payments while keeping its private keys offline the entire time, per Bitcoin Magazine’s coverage.

The trick is atomic swaps into Liquid, Blockstream’s federated Bitcoin sidechain. Incoming Lightning payments convert automatically into Liquid bitcoin and settle into the Jade-secured wallet; outgoing payments reverse the process, with the Jade device signing the transaction before funds ever leave cold storage. Blockstream chief product officer Jeff Boortz called it “a breakthrough for self-custody,” adding that “Jade is the first hardware wallet in the world to send and receive Lightning payments while keeping your keys fully offline.” Chief marketing officer Peter Bain put the practical benefit more simply: “the result is faster payments, stronger self custody, and fewer unnecessary transactions.”

This matters for the institutionalization story precisely because it cuts against it. Every custody-grade product described above solves Lightning’s self-custody friction, constant uptime, channel management, hot wallet exposure, by asking the user to trust a third party instead. Jade’s approach solves the same friction without that tradeoff, at the cost of being slower and more manual than an API call. It is a reminder that the custodial wave is a choice enterprises are making for convenience, not the only path available.

Banks Go Lightning Native: Xapo and Nubank

The 2026 product wave did not emerge from nowhere. It was built on integrations that licensed banks and fintechs had already been running for two or three years. Xapo Bank, a Gibraltar-regulated credit institution, became the first fully licensed bank to enable bitcoin deposits over Lightning through a partnership with Lightspark, letting members deposit and withdraw instantly instead of waiting on on-chain confirmations and the fees that come with them.

Nubank, Latin America’s largest digital bank with more than 100 million customers across Brazil, Mexico, and Colombia, struck a similar partnership with Lightspark in June 2024 to integrate both Lightning and Universal Money Addresses, a protocol for routing payments across different networks using a single identifier, according to the companies’ joint announcement. Nubank Cripto executive director Thomaz Fortes called it part of the bank’s mission to deliver services “with greater speed and lower costs through blockchain technology,” while Lightspark co-founder and CEO David Marcus said he was excited about “playing a role in bringing Lightning to Nu’s 100 million customers.”

Two or three years on, those integrations look less like early experiments and more like the reason 2026’s custody, credit, and yield products had somewhere to plug in. A bank that already routes retail deposits over Lightning is a natural customer for a Lightning-native credit line or yield product once one exists.

The Merchant Layer: Square, Block, and a Wall Street Debut

The other end of the pipe is where people actually spend the money. Block, the Jack Dorsey-led company behind Square, Cash App, and Tidal, began onboarding Square sellers to accept bitcoin payments over Lightning starting November 10, 2025, with zero processing fees promised through 2027, rolling out to roughly 4 million US point of sale merchants, according to Decrypt’s coverage. Merchants accept bitcoin directly through existing Square hardware, with Lightning handling settlement so a coffee shop transaction clears in seconds rather than waiting on a block confirmation.

The appeal for a merchant is narrower than it sounds: card networks charge interchange fees that eat into thin retail margins, and on-chain bitcoin settlement can take minutes and cost more in fees than a cup of coffee is worth. Lightning solves both problems by settling in the background while the customer walks out with their order, the same basic pitch Strike, Cash App, and a handful of exchanges have made for years. What is different in 2025 and 2026 is the scale of a single rollout landing on 4 million merchant accounts at once.

The rollout picked up fresh attention in mid-2026 for an unrelated reason: Block joined the S&P 500, with its stock climbing more than 14 percent over the following week as index inclusion drove passive buying, the same stretch in which the company was onboarding its first live bitcoin-accepting sellers. The coincidence mattered symbolically. A company processing Lightning payments for millions of small merchants was, at the same moment, being absorbed into the benchmark index that defines mainstream American equity markets.

None of this means Lightning payments are common at the register yet; zero fee pricing through 2027 is as much a customer acquisition subsidy as a sign of organic merchant demand. But it does mean the distribution is now in place at a scale no earlier merchant integration had reached, sitting alongside the custody and credit products further up the stack.

The 2026 Institutional Lightning Stack at a Glance

Laid out together, the past eight months add up to a recognizable stack, running from custody at the bottom to consumer-facing payments at the top.

Two things stand out. First, most of this stack is less than eight months old, even though the bank-level integrations at the bottom of the table go back to 2023 and 2024. Second, nothing on this list is a pure self-custody retail product; that territory now largely belongs to channel-less alternatives like Ark and Spark, which trade off differently against Lightning and sit outside the scope of this piece.

Where the Capacity Is Actually Coming From

The capacity trend underneath all of this is choppier than the product headlines suggest.

At bitcoin’s price in mid-July, in the mid 60,000 dollar range, that May reading alone is worth somewhere in the neighborhood of 300 million dollars, and even that likely understates the real total. Spark’s researchers estimate private and unannounced channels, the kind used by mobile wallets and the custody products described above, hold at least twice the publicly visible amount. Separately, Coinbase disclosed that more than 15 percent of its bitcoin withdrawals were already moving over Lightning by mid-2025, a sign the shift toward institutional rails was underway well before this year’s product wave, per Spark’s research.

Put the two threads of this article side by side and the institutional products described above are still a small fraction of total public capacity: BitGo’s initial Lightning Earn seed was 10 BTC against a network-wide total near 4,898 BTC, and even generous estimates of Voltage’s and Amboss’s combined managed liquidity would not move that headline number by much yet. The story in mid-2026 is less about institutions already dominating Lightning’s capacity and more about the rails being built for that to happen if and when larger allocators follow.

The Regulatory Backdrop: the GENIUS Act’s July 18 Deadline

None of this was built in a regulatory vacuum. The GENIUS Act, the federal law governing payment stablecoins, was signed on July 18, 2025, and gave its six implementing regulators, the OCC, FDIC, NCUA, Treasury, FinCEN, and OFAC, exactly one year to finalize rules, a deadline that lands on July 18, 2026, two days after this piece was published. The OCC published its main proposed rule in February, added a follow-up anti-money-laundering proposal in June, and by mid-July was drafting final text against the clock alongside the other five agencies.

The stakes run directly through this article. USDT increasingly moves over Lightning channels through Taproot Assets, and the same OCC that conditionally approved BitGo, Circle, Ripple, Fidelity, and Paxos for trust bank charters is now finishing the rulebook that will decide which of those companies, and which of their competitors, can legally issue or custody a payment stablecoin for US customers. Circle, notably, converted its own conditional approval into a final one on July 10, 2026, clearing it to launch Circle National Trust, according to CoinDesk’s reporting, just over a week before the statutory deadline. Circle CEO Jeremy Allaire called the approval “a defining step in bringing blockchain technology and digital assets into the core of the U.S. financial system.” For a deeper look at how enforcement and rulemaking fit together across the SEC and CFTC, see our explainer on 2026 crypto enforcement.

None of the products described in this piece are stablecoin issuers themselves. But BitGo, Voltage, and Amboss are all building infrastructure that regulated dollar tokens increasingly ride on top of, which means the finished GENIUS Act rulebook will shape which institutions are comfortable plugging into Lightning at all.

The Custody Paradox: What Retail Still Gets to Keep

It is worth stating the obvious tension plainly. Lightning, like Bitcoin itself, was built around the idea that users hold their own keys and trust code rather than institutions. Nearly everything described in this piece asks users to do the opposite: trust BitGo with custody, trust Voltage with channel operations, trust a lending desk’s underwriting. That is not a criticism so much as a description of what institutional adoption actually means in practice. Enterprises do not want to run nodes any more than most people want to run their own mail server; they want an API and someone accountable on the other end of it.

The concentration this produces is real and worth tracking. Spark’s research already shows capital and routing volume consolidating into fewer, larger, better-run nodes rather than spreading across thousands of small operators, a dynamic with an obvious parallel on Bitcoin’s base layer, where mining hashrate has followed a similar path toward a handful of dominant pools, a concentration this outlet has covered in detail in our explainer on who actually controls Bitcoin’s hashrate. Custodial Lightning liquidity is not identical to mining pool concentration, but the underlying pattern, capital-intensive infrastructure consolidating around a few well-capitalized operators, rhymes. It is also worth noting that concentration in custody does not automatically mean concentration in routing; a custodian like BitGo can in principle deploy client liquidity across many independent node operators rather than running all of it through a single choke point, though how any given provider actually structures that is rarely disclosed in detail.

None of this changes Lightning’s base security model. Channels still settle through the same hashed timelock contracts whether a custodian or a hobbyist opened them, and the protocol-level risks this outlet has covered before, from channel jamming to the long-run quantum computing question, apply regardless of who holds the keys. What changes is a different, more familiar kind of risk: whether the custodian is solvent, whether the credit line is well underwritten, whether the company holding your Lightning Earn deposit stays in business. Retail users who want Lightning-speed payments without any of that can still run a self-hosted node, or use a channel-less alternative built for lighter operational overhead, though that territory is its own story.

What’s Next for Institutional Lightning

A few questions will determine whether this wave keeps building through the second half of 2026:

• Whether more licensed banks follow Xapo and Nubank once the GENIUS Act rulebook is finalized and stablecoin issuance rules are settled, since several of the newly chartered trust banks have stablecoin plans of their own
• Whether Lightning-native credit and yield products stay concentrated in a handful of providers like BitGo, Voltage, and Amboss, or attract competition now that the model is proven
• Whether autonomous payments become a second major growth driver alongside enterprise treasury use cases, as toolkits emerge that let AI agents pay each other over the network
• Whether public capacity growth eventually catches up to the product layer, or whether institutional Lightning keeps growing mostly through private, unannounced channels that never show up in the public statistics

Our coverage of the agent payments push goes deeper into how that front overlaps with the institutional buildout described here, since the same custody and compliance rails enterprises want for their treasuries are also what agent payment infrastructure needs to avoid becoming an unaccountable black box.

The honest summary is that Lightning in mid-2026 is running two parallel tracks at once. Public capacity and node counts tell a story of a maturing but not obviously booming network. The product layer being built on top of it tells a much more aggressive story: a freshly public custodian, a Gibraltar-licensed bank, a hundred-million-customer fintech, and Jack Dorsey’s payments company all treating Lightning as infrastructure worth building enterprise products on. Whether that enterprise layer eventually pulls more transaction volume onto the base network, or simply becomes a faster, better-capitalized way to move dollars that happens to settle through bitcoin, is the open question for the rest of the year.

Frequently Asked Questions

What is the Lightning Network?

The Lightning Network is a second layer payment protocol built on top of Bitcoin. Instead of settling every transaction on Bitcoin’s base blockchain, users open payment channels, move funds back and forth off-chain as many times as they want, and settle the net result on-chain only when a channel closes. That design allows near-instant, low-fee bitcoin payments that would be impractical to run directly on the base layer, and it is the same infrastructure now being wrapped in custody, credit, and banking products.

Why are institutions like BitGo and Block suddenly building on Lightning?

Three things converged in 2026: clearer US regulation, including OCC trust bank charters and the GENIUS Act’s stablecoin rules; more mature infrastructure, including custody grade tooling from providers like Voltage and Amboss; and straightforward economics, since Lightning settlement is markedly cheaper and faster than moving bitcoin on-chain or through card networks. Together these made it viable for regulated companies to offer Lightning-based products without operating the underlying nodes themselves.

Is it safe to keep bitcoin in a custodial Lightning product like BitGo’s Lightning Earn?

It depends what risk you are trying to avoid. Custodial Lightning products remove the technical burden of running a node and managing channel liquidity, but they reintroduce counterparty risk, since your bitcoin sits with a custodian rather than in a wallet you control. That can be a reasonable tradeoff for an institution that wants routing fee yield without operational overhead, but it is a different risk profile than self-custody, and any yield product’s custodian, underwriting, and redemption terms deserve the same scrutiny you would apply to a bank.

How does BitGo’s Lightning Earn actually work?

Institutions deploy a minimum of 1 BTC as Lightning liquidity through infrastructure BitGo licenses from Amboss. The deployed bitcoin either sits in payment channels and earns a cut of routed payments, or gets leased as capacity to other network participants who need it, with Amboss’s automation handling peer selection, fee setting, and channel sizing. BitGo retains custody of the underlying bitcoin throughout, and returns are paid in bitcoin rather than a fiat-denominated yield.

How does the GENIUS Act affect Bitcoin’s Lightning Network?

The GENIUS Act itself regulates payment stablecoins, not Bitcoin or Lightning directly, but the two are increasingly intertwined because tokens like USDT now move over Lightning channels through Taproot Assets. Six federal regulators had until July 18, 2026, one year after the law’s enactment, to finalize implementing rules covering reserves, redemption, and who can legally issue a payment stablecoin to US customers, rules that will shape which custodians and payment companies are comfortable running dollar-token rails on top of Lightning infrastructure.

By the HOGE Wire markets desk.