Crypto cards with a trading desk in the app

As of October 2026, outside the exchange-issued cards, 12 of the 141crypto cards you can get let you open a leveraged position in the same app that issues the card. 9 of them execute on Hyperliquid, which makes them 9 front ends onto one order book. All but two of them name the venue in their own documents. Nobody had added them up.

A crypto card used to be a spending accessory. Over the past year a different shape has appeared: the wallet app that issues the card also carries an order book, and the same login that pays for coffee can open a forty-times-leveraged short. We screened the 141 cards obtainable on 4 August 2026 and set aside the ones issued by exchanges, where a derivatives desk is the parent company’s main business and tells you nothing. We found nine built that way. A second look on 27 October 2026 added three: Bitget Wallet and BloFin Wallet, whose apps send perps to Hyperliquid rather than to the parent exchange, and Bitpanda, which runs its own spot-margin book. That makes 12 of the 141 cards you can get today. The exclusion is our judgement about what is worth reporting. The data applies no such filter, and the group it removes has no published size: no column in the dataset records issuer type, so any count of it would be a brand-name guess.

The finding is narrower than “wallets are becoming exchanges”: 9 of the 12 route to the same venue, so what the cardholder picks is a front-end onto one market.

9 out of 12 is no surprise by itself. Hyperliquid carries the deepest on-chain perpetuals book, and an app adding a desk routes to the venue that can fill the order. Each time it is the obvious call, and better for the trader than a thin book of the wallet’s own. The sum is one unlabelled dependency behind 9 products a reader would name as different companies.

One venue, two country lists, and a transfer nobody advertises

1. 9 of 12 desks execute on Hyperliquid: Hyperbeat Card, Based Card, Tria Card, MetaMask Card, Ledger Card, xPortal Card, Coin98 Fusion Card, BloFin Wallet Card and Bitget Card. These are interfaces onto one order book. A halt or oracle failure there reaches every one of them. Only Tria and Bitget Wallet have somewhere else to go: Tria ships a second route on Decibel, and Bitget Wallet one on PopDEX.
2. The same 9 are running someone else’s venue, and two of them never name it. Ledger, MetaMask, Coin98 and Based name the venue on the same page as the feature; xPortal and Tria a line or a page away; Hyperbeat in its Terms, which define perp trading as executing “from your Pay Wallet, directly on Hyperliquid”. BloFin Wallet and Bitget Wallet do not name it on any page we could read. BloFin’s help article calls the desk “an on-chain DEX perp experience”, and the Hyperliquid link comes from the code in its shipped app. Bitget’s comes from Hyperliquid’s own builder-fill data, matched to Bitget Wallet by a third-party label. Nobody had added the dependency up, which is why the count is the finding and not any single disclosure. The three that run their own book are Jupiter, on its own Solana program, Revolut, through a brokerage authorised by the Bank of Lithuania, and Bitpanda, which lends the margin position itself.
3. “Trade and spend from one account” is true of none of them. Every app keeps the margin in a different account from the balance the card spends, whatever the landing pages imply. Hyperbeat was the exception on this page for seven weeks, on the strength of its terms, until its own funding guide described the transfer in both directions on 28 September 2026: money goes in from the spending account and comes back with a Sweep to Card button. The one-account language it does publish covers spot and perps, not the card.
4. The card’s country list is not the desk’s country list. Ledger blocks its desk in five jurisdictions (the UK, the US, Ontario, France and Belgium) and sells the card in all five, so a holder in any of them gets a card and no desk. Two of the five are EEA states, which is easy to miss when the card’s row says only “EEA”.
5. 12 candidate groups were rejected, and the reasons are published below. The two largest rejections are whole classes: collateralised borrowing dressed as leverage, and tokenised shares held at 1:1.

This page counts where the orders go. What to check card by card before you open a position, in the order that matters (reach, country, cost), is in our trading desk guide.

The 12 cards

On all 12 the trading margin sits in a separate pot from the card, and money has to be moved across before it can back a position. Hyperbeat was the last card listed as sharing one balance, until its own funding guide showed the transfer step on 28 September 2026.

Where the orders go

The chart counts front-ends per venue. Tria and Bitget appear twice because each ships two routes, which is why the bars sum to more than 12.

We have written this shape before, in the BIN-sponsor concentration research: several consumer brands, one regulated entity underneath, and a failure mode that only becomes visible on the day it fires. The difference here is that a BIN sponsor failing stops your payments, while a venue failing strands an open position with money in it. This venue has already fired three times; the dates are in the next section.

What is Hyperliquid?

Hyperliquid is the venue 9 of these cards route to, and none of them runs it. It describes itself as a layer-one blockchain, not a broker: its own docs call it “a performant blockchain built with the vision of a fully onchain open financial system”, running a custom consensus called HyperBFT, with the order books living in a component called HyperCore where “every order, cancel, trade, and liquidation happens transparently with one-block finality”. There is no licensed intermediary in that sentence, which is why the compensation schemes and the negative-balance rule that come with Revolut’s brokerage have no equivalent here.

It also means the costs and the failure mechanics are published once and apply to all 9, whatever the card app is called. Perps fees start at 0.045% taker and 0.015% maker and fall with rolling 14-day volume. Funding is charged every hour, paid between traders rather than to the platform, with the interest component fixed at 0.01% per eight hours. That is 11.6% a year, paid by longs to shorts before the premium component moves it either way. The card app can add a fee of its own on top, through what the venue calls builder codes: Coin98 publishes 0.04%, which takes the entry cost from 0.045% to 0.085%. That is nearly double, and it is disclosed in the docs and nowhere near the app screen.

Those percentages are quoted on the position, and the reader’s money is the margin. At the 40x ceiling several of these apps advertise, a position is 40 times the margin behind it: the 0.045% taker fee is 1.8% of your margin each way, about 3.6% for the round trip before the market has moved at all, and the fixed part of funding alone runs 1.2% of margin a day (0.01% of the position every eight hours, three times over).

What each front end adds on top

Coin98 is the clear case. Of the 9 front ends on Hyperliquid, two publish the add-on as a number (Coin98 and Based). Three publish only an all-in rate, from which the add-on can be worked out: MetaMask’s comes to 0.10% at its entry Rewards level, the largest in the set, BloFin Wallet’s to about 0.055% on our arithmetic, and Bitget Wallet’s to 0.045%. Both MetaMask’s and Bitget’s also show in Hyperliquid’s own builder-fill data. Tria’s documents give three different answers and no number; its app showed 0.05% on 27 October 2026, and the builder-fill data shows the same rate. Hyperbeat, Ledger and xPortal publish nothing at all, though Hyperbeat’s builder fee is visible on chain: 0.025% of notional on every fill in Hyperliquid’s builder-fill data for October 2026.

The liquidation numbers are what the leverage ceilings on the front pages mean in practice. Maintenance margin is half the initial margin at maximum leverage, which the docs put between 1.25% on 40x assets and 16.7% on 3x assets. On a 40x position, a 1.25% move against you starts the process. What happens next depends on the order book: if market orders close the position and bring the account back above maintenance margin, whatever collateral is left stays yours. If they cannot, and equity falls below two-thirds of that threshold, a backstop liquidation moves the position and the cross margin to a liquidator vault, and the docs say plainly what the trader is left with then: “if the trader has no isolated positions, the trader ends up with zero account equity”. That last clause is also the argument for using isolated margin rather than cross for a position you are not watching. Isolated margin confines the loss to the margin behind that one trade, and it is a setting in the app, not a property of the card. If an account still goes negative, auto-deleveraging closes positions on the winning side at the previous mark price to keep the platform solvent. So “a liquidation costs you the margin” is the good case, and a profitable position being closed for you is a documented outcome.

The venue has a record to check this against. On 26 March 2025 the venue’s validators voted to delist the JELLY market and every open position in it was settled. That is the documented behaviour: “when an asset is delisted, all positions are settled and open orders are cancelled”. The venue then announced that long holders would be refunded “as if their position settled at the closing price of 0.037555”, which is itself the admission that the actual settlement went against them. On 10 October 2025 the first cross-margin auto-deleveraging in the venue’s history closed positions on the profitable side. That account of it comes from the founder’s personal post, not the venue’s own channel: “during recent volatility, Hyperliquid had 100% uptime with zero bad debt. This was Hyperliquid’s first cross-margin ADL in more than 2 years of operation”. Order entry has also failed outright: the status page records 37 minutes on 29 July 2025 when “orders were delayed in being sent to the nodes”. A card app cannot route around any of it.

These three incidents do not establish a loss to anyone reading this. The March and July ones came before MetaMask switched its perps on, on 8 October 2025 by its own announcement. The October auto-deleveraging came two days after, so we cannot say that no holder of these 12 cards had a position at risk that day, and we have not checked whether any did. Hyperbeat’s and xPortal’s perps are 2026 products. What the incidents show is how the venue behaves under stress. We also have not computed an uptime denominator: three incidents over roughly a year and a half is a rate we have not measured against anyone else’s.

Is the card balance the trading balance?

No, on all 12. This decides how far a bad trade can reach, and the marketing blurs it. “All from the same account you save and spend from” is Hyperbeat’s own line, and its terms still define both card spend and perp trading against the same Pay Wallet. Its own funding guide says otherwise: “Before you can trade, you need funds in your trading account”, with a Sweep to Card button for the way back. The contradiction is Hyperbeat’s. We follow the guide, because that is the screen a cardholder uses.

The other 11 document a split themselves, and three give it away in the small print: a minimum deposit to open the trading account. MetaMask asks $10 into a Hyperliquid account, Coin98 5 USDC into a dedicated futures account, Tria $10. A minimum deposit is a reliable tell, because money that has to be moved was never in one place. Based says it outright instead: card and trading wallets cannot be converted into one another, and card withdrawals are not enabled. Ledger, xPortal and Jupiter show it from the card side: each says the card spends a balance held by a card partner, which you fund by transfer from the wallet. Bitget Wallet’s help centre says it plainly: “Your Perps account is separate from your wallet account.” BloFin Wallet funds its card by a deposit step into a balance the card programme holds. Bitpanda moves the collateral into a margin wallet whose assets “cannot be withdrawn, swapped, or used elsewhere”, and the card spends only the main and fallback payment wallets. Revolut needs no tell at all, because the separation there is legal before it is a product decision. Its CFD cash is client money, which Revolut says is segregated and safeguarded, with the caveat it prints in the same breath: up to 20% of that value can sit on accounts inside Revolut Group, and only the remainder goes to a third party.

Based looks like an exception and is not one. Its web terminal offers a “Unified Account” where “a single balance per asset covers everything: perps, spot, and cross margin”. That is a setting on the trading side, with no bridge to the card. The FAQ is where the card question gets answered, as set out above.

“Self-custody card” often describes the wallet rather than the money already loaded to spend. Six issuers here pair a self-custody wallet with a card balance someone else holds. Coin98’s card balance sits with DCS. BloFin Wallet and Bitget Wallet each fund the card by a top-up into a balance held by the card programme. Ledger’s FAQ says funds moved to the Baanx-run card account “are no longer under self-custody”. xPortal’s card takes only EUR deposits into an e-money balance. Jupiter says card balances “are managed by the issuing partners”, Rain or DCS. Tria’s card is a 0% APR credit line against collateral, with no debit balance, repaid automatically from the self-custody wallet.

The two country lists

Three of these cards publish a coverage map for the card and a different one for the desk, and one more card was cut from the research when the two maps turned out not to overlap anywhere. Quoting one number for both is how a comparison page tells a reader they can trade when they cannot.

The table below carries 7 rows because several different problems live in the same field. Ledger and Revolut sell card and desk to different countries, and Jupiter may: its card now takes US and Singapore residents, whom the app terms we read in August barred. Hyperbeat and Tria are a second kind of trouble: it is unclear whether a US resident can get the card at all. Hyperbeat’s own documents disagree with each other on the point. Tria keeps a separate US programme under its own card terms, and no Tria page says that programme is taking applications. BloFin Wallet sells its card in 48 listed countries and publishes no list for the desk, so the overlap cannot be checked. Bitpanda is the mild case: its card is sold in the eurozone, which sits inside the EEA-and-Switzerland area where its stock margin opens, and its crypto margin publishes no country list.

Where each desk is closed, in the issuer’s words

Four issuers publish a desk restriction (Bitpanda only for its stock margin), one publishes it only by reference to sanctions regimes, and seven publish nothing we could find. Restrictions belonging to a different product of the same issuer are excluded here on purpose: Hyperbeat’s 24-country list is for its raffle product, Based’s 92-country list is for the Visa card, and Jupiter’s one region sentence is about prediction markets.

Three run their own book, one is a regulated CFD broker

Three entries run their own book: Jupiter, on its own Solana program, Revolut, and Bitpanda. Only Revolut trades under a licence written for the job. Its CFDs run through Revolut Securities Europe UAB, authorised by the Bank of Lithuania, at rates it says it generally offers: 1:20 on major indices and on gold, 1:10 on other commodities, 1:5 on single stocks. Negative-balance protection comes with that regime, and the same page states it: the margin-close-out rule “makes sure your account doesn’t fall into a negative balance”. It applies because the licence requires it for retail clients, which makes it firmer than a promise a firm writes for itself. Bitpanda is a licensed platform too, but its margin product is spot-based and runs on its own terms: crypto losses stop at the collateral under those terms, while on stocks its help centre says you may still owe money if the closing proceeds fall short. It publishes up to 10x on crypto and 20x on stocks, ETFs and ETCs. The 10 on-chain desks carry no equivalent and no compensation scheme. The seven of them that publish a ceiling at all run from 40x to 250x. Those are a different instrument from a 1:5 CFD on a single stock, and the figures are not multiples of one another.

The low rates come from the regime and say nothing about the product. They are the reason these 12 do not belong in one undifferentiated table.

What we left out

12 candidates or candidate classes were cut, and the reasons are published because most of them are cards a reader would expect to find here.

Arguments against this finding

The denominator is ours. 12 of 141 is a rate over a population we defined, and the group we took out, the exchange-issued cards, is the group that would dilute the finding, because every exchange runs its own book. We still think the cut is right, because an exchange having derivatives is not news. It is a judgement all the same. The size of the excluded group is unpublished because no column in the dataset records issuer type, and a reader is entitled to treat the absolute number as the sturdier one. The closest calls under our own rule were Bitget Wallet and BloFin Wallet, and both are now in: their parent companies run exchanges, but the app a cardholder installs sends perps to Hyperliquid, not to the parent’s book. Cards issued by the exchanges themselves, BloFin’s own exchange card among them, stay out.

Routing to the deepest book is also good for the trader. A wallet that stood up its own thin order book would fill worse and slip more. Each of the 9made the call that serves its own users. The aggregate describes where the liquidity is, and nobody behaved badly to produce it.

Almost nobody holds two crypto cards for trading redundancy. People hold a second card because the first one gets declined abroad. The diversification this page says you do not have is one hardly anyone was looking for. The argument survives in a narrower form: if you hold two and assume the desks are independent, they are not.

No harm to anyone here is on record either. The three incidents are real and dated, two of them predate MetaMask’s perps launch, the earliest Hyperliquid desk launch here we have a date for, and we have no record of a cardholder in this dataset losing money to venue concentration. It is a tail risk written down before it fires, which is both the reason to publish it and the reason to keep it in proportion.

None of that moves the count. 9 of the 12 still execute in the same place, and none of them says how many of the others do.

Reader questions on scope, liquidation and coverage

Methodology

We started from every card in the Sweepbase dataset that a person can obtain today (141 of them) rather than from a shortlist, because the prose in our own database is not a usable signal here. When we screened in August the word “trading” appeared on 16 live rows, seven of them exchange cards whose desk was never in question. On 27 October it appears on 21, and it is missing from ten of the 12 rows that do belong here. Matching on it would have produced a different list, wrong in both directions.

Each card was checked against its issuer’s own site, docs and help centre, and every inclusion carries a verbatim quote with the URL it came from. Claims that only third parties made were rejected: a claim about an issuer has to come from the issuer.

Two things had to be true of every row: the quote has to exist on the page cited, and that same page has to carry the figures the row publishes. A real sentence attached to a page that does not support the row is still a sourcing error. The first nine rows were read three times against that standard, each time from a clean start, and the three added on 27 October 2026 against the same standard.

We have opened a position on none of these desks, so every fee, liquidation and balance claim here comes from documents rather than from trading. We have not installed all 12 apps either. We have looked inside two: Tria, whose card the author holds, and Jupiter, where the Perps tab was checked in an account we own. The other 10 are described from what the issuer publishes about the screen, and for BloFin Wallet and Bitget Wallet also from the app’s shipped code and the venue’s own builder data.

Corrections

Six corrections were made on 5 August 2026.

The largest is Ledger’s geography. This page said the desk names five restricted jurisdictions and the card is sold in two of them. It is sold in all five: France and Belgium are EEA states, Ontario is in Canada, and the card’s own row lists the EEA and Canada. So the group that gets a card and no desk is larger than we printed, and the error ran against this page’s own argument rather than for it.

The second is an attribution. We wrote that the venue’s own account posted only that uptime held, while the description of the October auto-deleveraging came from the founder’s personal account. Both statements sit in the same post from that personal account. The facts are unchanged; the split was ours.

Then three numbers. The fixed part of funding was given as 1.3% of margin a day and it is 1.2%, because our figure re-compounded a rate the same paragraph had already called simple. Jupiter’s row had positions collateralised by the JLP pool, which is backwards: the trader posts the collateral and borrows from the pool. And “trading” appears on 16 rows of our database, not 15. It became 16 the day an unrelated edit added the word to Revolut’s countries field, which is the argument against matching on prose in one line.

Last, Revolut’s client money. We described CFD cash as sitting in a segregated client-money account at an eligible third party and stopped there. Revolut’s own legal page carries the rest of the sentence: up to 20% of that value may be held on accounts inside Revolut Group. The omission flattered the one card here we hold up as the legally separated case.

A re-read of the card pages on 25 September 2026 made one more. Ledger, xPortal and Jupiter were listed as not stating whether the card balance and the margin are separate. All three do say it, from the card side: the card spends a balance held by a card partner, which you fund by transfer. They now count among the cards that document the split. The same re-read found that the BloFin card we had cut as a wallet card is the exchange’s own, which leaves Bitget as the one close call.

The same day we corrected two statements about the venue incidents. The page said MetaMask’s perps were a 2026 product and that no cardholder here had a position at risk on any of the three days. MetaMask announced its perps as live on 8 October 2025, two days before the October auto-deleveraging, so that claim could not stand. Separately, the page said the author’s Tria card was one of two here with an affiliate arrangement; it is one of four.

The biggest correction is the newest. From publication until 28 September 2026 this page said Hyperbeat kept card spending and trading margin in one wallet, and it was the only card here we said that about. The evidence was Hyperbeat’s terms, which define both against one Pay Wallet, and the absence of any transfer step in 434,000 characters of its documentation. The entry named the disproof it would accept, and Hyperbeat has since published it: a funding guide that opens “Before you can trade, you need funds in your trading account”, lists the spending account as a deposit route, and gives the money two ways back, Transfer and Sweep to Card. The terms have not changed since 1 March 2026 and still say Pay Wallet, so the contradiction is the issuer’s; we follow the guide, because the guide is what a cardholder operates. No card on this page now shares one pot, and the count that used to read one is zero.

Three figures in the fee and geography columns also moved that day. MetaMask has taken its perps fee schedule off the page entirely, so the combined rates we printed, and the comparison we drew from them against the venue’s own 0.045%, no longer have a source. That was wrong: the schedule had moved to the help centre, which on 27 October 2026 still gave 0.1432% taker at the entry Rewards level, and it is back in the table above. Tria’s own launch post says it charges a fee on top of the venue’s and names no number, where its futures terms say it charges nothing. Revolut’s “annualised 3% mark-up” on overnight financing could not be re-found on either document that names it.

On 27 October 2026 the list grew from nine to twelve, and the share on one venue from seven of nine to nine of twelve. Bitget Wallet had been left out because its parent runs an exchange; Hyperliquid’s builder-fill data, under an address a third party labels Bitget Wallet, shows its perps going to Hyperliquid, so the app passes the test this page applies. BloFin Wallet’s card joined the catalogue the same day, and its shipped app points to Hyperliquid too. Bitpanda came in with its own spot-margin book. The same day Tria’s app showed a 0.05% Tria fee on an order on its Hyperliquid account, a number none of its documents gives.

Sources: issuer sites, product documentation and help centres, linked and dated on every row of the card table above. The rows were read in August 2026, the card side of Ledger, xPortal, Jupiter and Coin98 again on 25 September, and all nine desks again on 28 September. The three rows added on 27 October were read that day, when the page was last revised. Card-level facts come from the Sweepbase dataset; scoring is explained in the rating methodology. Spot an error? Report a correction.

About the author: Mihail B. is the founder and editor of Sweepbase. He holds the Tria Card, which appears on this page and is one of five cards here that Sweepbase has an affiliate arrangement with. The others are Hyperbeat, MetaMask, Ledger and Bitget. This page carries no affiliate links; the category listing does. How we verify is described on how we test.