How People Actually Use Lightning: Wallets, Zaps, and Tips

Every few days, another headline reduces Bitcoin’s Lightning Network to a single chart: public capacity up, node count down, monthly payment volume climbing toward a round number. With bitcoin trading around $64,700 as of this week, according to CoinGecko, those charts matter, but they describe an aggregate, not an experience. Nobody opens a capacity chart to send five dollars to a friend or tip a stranger’s post.

What actually decides whether Lightning feels usable is the layer underneath the stats: the wallet a person taps, the custody model behind it, and, increasingly, a separate but related phenomenon that has nothing to do with exchanges or stablecoins at all. On Nostr, the decentralized social protocol, tipping a post in sats has become one of the most active, organic things people do with Lightning, a practice with its own name, its own technical spec, and its own small economy. This is a look at both halves: how ordinary wallets actually route payments in 2026, and how zaps turned a social network into a live demonstration of what Lightning was built to do.

Custodial, Self-Custodial, or Something in Between

Every Lightning wallet sits somewhere on a spectrum, and where it sits determines almost everything about how it feels to use. At one end are fully custodial apps: the provider holds the balance, manages every channel, and the user just sees a number go up or down, the same mental model as any banking app. At the other end is running a full node with manually opened channels and self-managed liquidity, the approach Bitcoin’s earliest Lightning adopters had no choice but to use.

Almost nobody who isn’t a developer or a dedicated hobbyist lives at that second extreme anymore. The middle of the spectrum, where a wallet holds the user’s keys but leans on a Lightning Service Provider for inbound liquidity and channel plumbing, is where most of 2026’s growth in usability has actually happened. Two technical shifts made that middle ground viable: just-in-time channel opens, where a wallet requests a channel from an LSP the moment a payment needs one instead of asking the user to pre-fund liquidity, and splicing, which lets a wallet resize an existing channel without closing and reopening it on-chain. That shift matters commercially too: an LSP can amortize the cost of managing liquidity across thousands of users, which is a big part of why wallets built on this model have kept fees low even as they absorb the operational complexity their users no longer have to see. Channel-based liquidity isn’t the only model in production either; shared-UTXO and statechain designs trade some trust assumptions for skipping per-user channel management entirely, though this piece stays focused on the wallet layer sitting on top of whichever base model a given app chooses.

The Custodial Fast Lane: Cash App and Strike

For sheer reach, nothing in the self-custodial world competes with Block’s Cash App. In November 2025, Cash App bundled a set of Bitcoin features into a release the company called Cash Releases: instant Lightning payments that draw automatically from a user’s dollar balance, even if they hold no bitcoin at all, a Bitcoin Map for finding nearby Square merchants that accept BTC, and a Bitcoin Back program paying rewards on eligible purchases. Cash App counts 58 million monthly users, which means a Lightning payment feature shipped there reaches more people in a single update than the entire public Lightning Network has nodes, by a wide margin. Miles Suter, Block’s Bitcoin Product Lead, framed the release plainly: “Bitcoin was created to be peer-to-peer cash, and Cash App is building tools to make it work as intended, fast, open, and borderless,” Suter said, as Bitcoin Magazine reported at the time.

Strike has taken a different route to the same audience. Jack Mallers built the app on remittances and Lightning-powered send and receive first, then spent 2026 pushing it toward becoming a fuller consumer finance product. On July 7, Strike launched what it calls volatility-proof Bitcoin-backed loans: customers can borrow against BTC collateral at up to 45% loan-to-value without facing a forced sale if the price drops, a structure the company says can absorb an 80% decline as long as payments stay current, built on a $2.1 billion credit line with Tether and reported by The Block. “No margin calls. No price liquidations. No matter how far bitcoin falls, your bitcoin doesn’t move,” Mallers wrote announcing the product. Asked about self-custody, he was direct about where Strike still draws the line in the app’s favor: “The best thing you can do to prove that your Bitcoin is your Bitcoin is withdraw to cold storage, right? And we offer that for free.”

Neither app asks its users to think about channels, liquidity, or routing fees, and between them, Cash App and Strike illustrate a pattern that shows up across every custodial Lightning product: the fastest path to real-world usage runs through an app people already have open for other reasons, not through a dedicated Bitcoin wallet competing for a new download.

Wallet of Satoshi and the Regulatory Squeeze

Wallet of Satoshi built its popularity on being the simplest possible on-ramp: no account, no seed phrase to write down at setup, open the app and receive sats. That simplicity became a liability once regulators started drawing lines around custodial crypto services. The app pulled itself from United States app stores in November 2024. Then, in January 2026, it blocked custodial service across the entire European Union, a shift CryptoTimes reported affected users from Germany to Poland to the Netherlands.

Two EU rules converged on the same window. The Markets in Crypto-Assets Regulation, known as MiCA, requires any custodial crypto-asset service provider to hold a national license, backed by own-funds requirements that scale with the service offered; for custody and administration specifically, the floor is €125,000 in minimum capital, per the European Securities and Markets Authority. Alongside it, DAC8, the EU’s eighth Directive on Administrative Cooperation, took effect on January 1, 2026, and requires custodial providers to identify their users and automatically report transactions to national tax authorities. The timing wasn’t arbitrary either: several EU member states’ MiCA transitional arrangements for firms already operating were set to close by mid-2026, so a wallet weighing whether to invest in a full license had a hard deadline concentrating the decision. For an app built around not asking for an account, both requirements cut against the product’s original design.

Rather than close entirely, Wallet of Satoshi is offering a self-custodial version built on Lightspark’s Spark network, the statechain-based layer covered in HOGE Wire’s Lightning, Ark, and Spark comparison, in both the US and EU. Users keep roughly the same interface; what changes underneath is who holds the keys, which moves the app outside the custodial licensing regime it could no longer economically satisfy for a free product.

It’s a useful contrast to how the United States has handled its own 2026 stablecoin deadline. Rather than a rule landing on schedule, the GENIUS Act’s one-year deadline for federal regulators to finalize stablecoin rules passed on July 18 with nothing final at any of the six primary agencies, a gap HOGE Wire covered in Lightning’s Dollar Rails Are Live. Washington’s Rulebook Isn’t. The EU, by comparison, already has both an operative licensing regime and an operative reporting regime, and a consumer wallet has already changed its product because of them. Regulatory speed cuts both ways: slower rules mean less near-term disruption, but also less clarity about what a compliant Lightning wallet is actually allowed to do.

Self-Custodial Without the Homework: Phoenix and Breez

Between the fully custodial apps and the run-your-own-node crowd sits a set of wallets trying to deliver both self-custody and a custodial-grade experience. Both Phoenix and Breez answer the same question, how to keep self-custody without asking a user to think about liquidity, but they answer it for different people. ACINQ’s Phoenix maintains a single dynamic channel per user and uses splicing to resize it automatically as needed, so on-chain and off-chain balances blend into one number on screen. The user holds the keys the entire time; they just never have to think about what a channel is. Fees come from two places, the on-chain mining cost of the occasional splice and standard routing fees for sends, rather than a flat custodial spread.

Breez targets merchants and creators who want point-of-sale functionality built in. Its current version runs on a Nodeless SDK over the Liquid sidechain rather than a directly managed node, branded as Misty Breez since its April 2025 relaunch, and charges a one-time fee of about 0.75% with a roughly 2,000-sat minimum, according to Spark’s wallet comparison research. Muun occupies a simpler niche again, routing Lightning sends through on-chain submarine swaps with no LSP relationship at all; the tradeoff shows up during fee spikes, when that same research notes users have reported paying 5,000 or more sats in mining costs for what should be a small payment, a reminder that removing the LSP layer removes a dependency but not the underlying cost of touching the base chain.

Running the Full Node: Zeus and the Power-User Path

Zeus sits at the opposite end of the spectrum from Cash App on purpose. It runs a full embedded LND node on the phone itself, or connects to a node the user already operates elsewhere, and hands over manual control of channel opens, fee policy, and liquidity management. There is no LSP softening the experience. That is exactly why the people who choose it tend to already understand what a channel is before they install the app.

This path is also, numerically, a shrinking share of the network. Public node count has fallen to roughly 17,400 as of the most recent published reading, down from a 2022 peak near 20,700, according to Spark’s ongoing state-of-the-network research, even as public capacity and payment volume have both grown. The two trends together tell a consistent story: capital and usage are concentrating into fewer, better-run endpoints, whether that is an exchange’s routing node or an LSP serving thousands of Phoenix and Breez users, while the number of individuals running their own node keeps thinning out. Running a node was never going to be how Lightning reached 58 million Cash App users; it remains how a smaller, technically committed group chooses to hold Lightning to its original self-sovereign premise.

Lightning Wallets Compared

Put side by side, the custody model each wallet chooses says more about who it’s built for than any feature list could. The table below groups the wallets discussed above by custody type, typical cost, and the user each is best suited to.

Nostr and the Rise of Zaps

Nostr is a minimal, decentralized protocol for social content, first written in 2020 by a pseudonymous Brazilian developer known as fiatjaf, built partly as a reaction to moderation disputes and technical frustrations with earlier decentralized social efforts. Instead of one company running the servers, anyone can run a relay, and clients read and write notes across whichever relays they choose. The protocol has grown well past its early-adopter phase, with well over a hundred clients and hundreds of independently operated relays now in production, though exact figures shift constantly since anyone can spin up a new relay without asking permission. Jack Dorsey, the former Twitter chief executive who now leads Block, became an early funder; fiatjaf later split part of that funding with William Casarin, a former Blockstream engineer who built Damus, one of the first Nostr clients to reach Apple’s App Store.

Casarin is also the person generally credited with building zaps, the mechanism that lets a Nostr user send a small Lightning payment directly to whoever posted a note, reply, or profile they want to reward. The spec that formalized it, NIP-57, was created in February 2023 and defines two linked event types: a zap request, generated the moment someone taps the lightning-bolt icon, and a zap receipt, published once the recipient’s Lightning wallet confirms the invoice was paid, as documented in the formal NIP-57 specification. Any client that understands both event types can show a running tally of sats a note has received, turning a like button into something with real value attached.

Zaps became a genuine flashpoint in 2023, when Apple briefly threatened to pull Damus from the App Store over its ability to send bitcoin tips, treating it as a digital-goods purchase subject to Apple’s in-app payment cut. Dorsey and Elon Musk both pushed back publicly, and Apple backed down. “Tipping on posts is not selling digital content. It’s a form of feedback,” Dorsey argued at the time, adding, “Why limit people sending bitcoin to each other? This is our one opportunity to build a truly global payment protocol for the internet,” comments NBC News reported as the dispute played out. That episode is years old now, but it’s the reason zaps carry cultural weight beyond their dollar value: a fight over whether a platform gets to tax a tip.

How Big Is the Zap Economy

Zaps are still a fraction of Lightning’s total payment volume in raw sats, but their activity is genuinely organic in a way a lot of Lightning usage isn’t. Volume surged past 20 million zaps in a two-week window in early 2025, according to Spark’s Nostr zap research, a pace that made social tipping one of the most consistently active things people actually do with Lightning rather than merely hold or trade it.

The client a person picks changes how much friction sits between wanting to zap something and it actually happening. Primal is the closest thing to an all-in-one option, since it ships with a built-in wallet, the Primal Wallet, that removes the setup step entirely; its version 3.0 release in March 2026 added zap-based poll voting alongside a remote-login feature. Damus, the client at the center of the 2023 Apple dispute, ships with no built-in wallet at all and expects a user to pair an external one over Nostr Wallet Connect. Amethyst, the leading Android client, sits in between, supporting the same NWC pairing model on a platform Apple never had leverage over in the first place.

Alby Hub has become something like the standard bridge between the two worlds: a self-custodial Lightning node that connects to web apps through Nostr Wallet Connect and doubles as a Lightning address for receiving zaps directly. It’s aimed squarely at the people most likely to receive zaps in the first place, creators, writers, open-source maintainers, and podcasters, rather than at casual senders, which is a useful signal about who Nostr’s Lightning economy is actually for: not traders, but people getting paid in public, one small payment at a time.

Value for Value: Podcasting’s Lightning-Native Model

Zaps grew out of social media, but the same NIP-57 mechanics, and the same underlying idea, power an older and in some ways more developed use case: Podcasting 2.0, an open set of RSS extensions built to route around ad-supported and subscription-locked podcast platforms. Its value for value model, championed early on by podcasting pioneer Adam Curry and developer Dave Jones through the Podcast Index project, asks listeners to pay creators directly for what they find worthwhile instead of routing everyone through a platform’s ad load or paywall.

Fountain is the clearest consumer product built on that idea. Listeners can boost an episode with a message attached to a Lightning payment, or turn on streaming sats that pay a creator continuously, a fraction of a cent at a time, for every minute they actually listen, a mechanic Fountain describes as a direct implementation of the value-for-value idea. A single popular episode can collect boosts from hundreds of listeners, each choosing their own amount rather than a platform-set subscription price, closer to how a busker’s guitar case works than to how a royalty statement works. Because there’s no platform intermediary skimming a cut, creators keep what listeners send in full. Fountain extended the model further with Fountain for Podcasters, a hosting service that distributes video alongside audio and supports the same streaming payments and subscriptions over open RSS, Lightning, and Nostr standards rather than a proprietary monetization layer.

What makes this specifically a Lightning story, and not just a creator-economy one, is the payment size. Streaming a fraction of a cent every sixty seconds is not something card rails can do; the processing fee alone would exceed the payment. Lightning’s near-zero marginal cost per payment is what makes the per-minute model mathematically possible at all, which is part of why value for value has stayed a Bitcoin-native niche rather than migrating to a card-based competitor.

Where AI Agents Fit In

Zapping a post and letting an autonomous software agent pay for an API call look like unrelated problems, but they share the same underlying constraint: both are payments too small and too frequent for card-network economics to tolerate. Lightning Labs built the L402 protocol specifically to let a piece of software request a Lightning invoice, pay it, and receive a token unlocking a resource, all without a human clicking a checkout button, and Google’s AP2 framework and Coinbase’s x402 protocol have since pushed similar machine-to-machine payment models from different angles. HOGE Wire’s deeper look at AI agents on-chain covers how that autonomous-payment layer is developing.

The connective tissue worth noticing here is that both use cases validate the same original Lightning pitch from a different direction. Zaps prove humans will use sub-cent instant payments when the friction is low enough; agent payments are a bet that software will use them even more, since a machine has no patience for a five-second card authorization and no objection to paying per request instead of per month.

The Security Trade-offs of Everyday Use

Every point on the custody spectrum trades one risk for another. A custodial balance on Cash App or Strike carries counterparty risk, the provider could be hacked, freeze funds, or simply make a mistake, but it also means account recovery doesn’t depend on safeguarding a seed phrase; losing a phone doesn’t mean losing the funds. A self-custodial wallet removes the counterparty but puts the entire burden of backup on the user. Losing the seed, or the encrypted channel backup a splicing wallet like Phoenix depends on, can mean losing funds permanently with no support line to call.

Browser-based and web-connected wallets add a further, more current risk. Alby’s browser-extension model, and any wallet paired over Nostr Wallet Connect, sit in the same category as browser-based Ethereum wallets when it comes to phishing exposure: a malicious site or a fake connect-wallet prompt doesn’t need to break Lightning’s cryptography, it just needs to trick a user into approving something they shouldn’t. HOGE Wire’s look at crypto phishing campaigns and the drainer economy covers how that social-engineering layer operates across chains, and none of its lessons stop applying just because the payment rail underneath is Lightning instead of Ethereum. A Lightning-specific variant worth watching for is address and LNURL spoofing, where a copied or manipulated payment string quietly redirects funds to an attacker instead of the intended recipient.

• Treat a self-custodial wallet’s backup, seed or channel state, exactly like a hardware wallet’s seed phrase: offline, never photographed, never pasted into a chat window.
• Treat a Nostr Wallet Connect string like an API key. Anyone who has it can spend from the wallet it’s paired to.
• Double-check a Lightning address or LNURL before a large payment; a single altered character sends funds somewhere else with no reversal possible.

The Regulatory Backdrop, and Where It’s Headed

Wallet of Satoshi’s EU exit and the GENIUS Act’s missed deadline aren’t the same story, but they sit on the same map. One targets custodial service providers directly; the other targets stablecoin issuers, which matters for Lightning mainly because dollar-denominated tokens increasingly ride Lightning’s rails too. Together they sketch how differently the two largest regulatory blocs are approaching money that moves like this. For the fuller picture of which US agencies police which corner of crypto, HOGE Wire’s mapping of US crypto enforcement in 2026 lays out the whole roster.

As CoinDesk noted in a one-year retrospective on the GENIUS Act published the day after the deadline passed, none of the six primary US agencies had a final rule in place despite a year of drafting. Neither approach is obviously right: the EU’s finished rulebook gives builders certainty at the cost of having already pushed at least one popular wallet to restructure its product, while the unfinished US rulebook gives builders room to experiment at the cost of not knowing which experiments will still be permitted once Washington actually decides.

What This Means Going Into 2026’s Second Half

Put the wallet layer back underneath the capacity charts and the aggregate numbers start to make more sense. Public Lightning capacity sits around 4,898 BTC as of the most recent published reading, spread across roughly 41,000 channels and those 17,400 nodes, per Spark’s research cited above, and that figure moves mainly when a handful of exchanges and institutional routing nodes adjust their positions, not when a new Cash App user sends their first five dollars. Retail usage, by contrast, is invisible in a capacity chart almost by design: a Cash App or Strike payment never touches the public channel graph the way a self-custodial payment does, since the provider nets flows internally wherever it can.

That split is likely to keep defining the network’s next phase. Growth in raw capacity will keep tracking institutional and exchange behavior, largely disconnected from how many humans are actually tapping send. Growth in actual usage, the kind that determines whether Lightning becomes a real payment habit rather than a curiosity, depends on three separate bets: whether custodial apps like Cash App and Strike keep shipping Lightning-powered features their users don’t even have to notice, whether LSP-backed wallets like Phoenix and Breez keep closing the gap with custodial UX without asking users to give up their keys, and whether the zap and value-for-value niche grows from a cultural signal into a revenue channel creators actually plan around. None of those bets show up on a capacity chart, at least not until they’ve already been won. Whichever pays off first will probably matter more to Lightning’s next chapter than any single capacity record.

Frequently Asked Questions

What is the best Lightning wallet for a beginner?

For most beginners, a custodial app already installed on their phone is the easiest starting point: Cash App or Strike both route payments over Lightning without requiring any setup. Readers who want to hold their own keys from day one without managing channels typically start with Phoenix or Breez, since both lean on a Lightning Service Provider to handle liquidity automatically.

What is a Nostr zap?

A zap is a Lightning payment attached to a post, profile, or comment on the Nostr protocol. It is defined by NIP-57, which creates a signed zap request when someone taps the lightning-bolt icon and a zap receipt once the recipient’s wallet confirms payment, so any Nostr client can show how many sats a note has received.

Is Wallet of Satoshi still available in the EU and US?

Not in its original custodial form. Wallet of Satoshi removed its app from United States stores in November 2024 and blocked custodial service across the European Union in January 2026, citing MiCA licensing and DAC8 tax-reporting rules. It now offers a self-custodial version built on Lightspark’s Spark network in both regions instead.

Do I need to run my own Lightning node to use it?

No. Most Lightning users in 2026 rely on either a custodial app like Cash App or Strike, or a self-custodial wallet such as Phoenix or Breez that automatically manages channel liquidity through a Lightning Service Provider. Running a full node with manual channel management, the approach used by wallets like Zeus, is largely a choice for experienced users rather than a requirement.

How do podcasters and creators get paid through Lightning?

Through a model called value for value, popularized by Podcasting 2.0. Apps like Fountain let listeners send creators direct Lightning tips called boosts, attach a message to a payment, or stream tiny per-minute payments while listening, with creators keeping the full amount since there is no platform intermediary taking a cut.

Written by the HOGE Wire markets desk.