Cards you can hold - for ATMs, hotels and the places where plastic still matters.
Short answer: KAST is our top-rated physical crypto card as of 2026-10-03 (7.1/10), Full KYC, 0% load fee.
cards on this page: 30 · issuing today (excl. watch list): 28 · independently verified facts: 73% · no KYC at first use: 3%
quick picks
The most complete stablecoin-spending account on the market: US/EU virtual bank accounts from anywhere, Visa Platinum cards with real 1.5% USD cashback on the free tier, instant cashback and no spending caps. Full KYC and a young operator are the trade-offs; the paid tiers only make sense for heavy spenders.
Europe's cleanest all-asset debit card: spend any of 600+ cryptos or even precious metals from one EUR card with zero monthly fees, 0% own FX markup, generous EUR 280k/month limits and an MFSA-regulated issuer. The 1% cashback only pays on crypto payments, ATM withdrawals cost 2%, and it is euro-area only.
The strongest mainstream exchange debit card for UK/EEA/US users on paper: free issuance, no monthly fee, 0% FX on the Mastercard rate, instant tiered cashback up to 2% and a metal card for high-balance users. The catch: cashback only starts above a GBP/EUR 200 average balance, crypto spend carries an undisclosed spread, and per-day limits are not published.
The cleanest bitcoin-rewards credit card structure documented: flat 1.5% in BTC uncapped with no annual fee, up to 4% for engaged users, issued by Celtic Bank under Visa. It is a credit product - VantageScore-based checks, bureau reporting, and variable APR 20.49-29.44% on carried balances.
The reference bitcoin-back debit card in the US: sats on every swipe via the Fold+ subscription (0.5% base plus rotating boosts up to 15%), FDIC pass-through on the USD balance via Sutton Bank, BitGo custody for bitcoin, Nasdaq-listed parent. The free tier earns no card cashback - Fold+ is the product.
The most mainstream way to hold BTC next to a Visa debit - but the card itself never touches bitcoin: every spend requires selling first, which is a taxable event per transaction. Use it for the Boosts and the frictionless BTC auto-investing, not for crypto-native spending.
The mainstream choice if you want crypto spend inside a full banking app - regulated, mature, multi-plan. Crypto card payments carry no exchange fee but the plan fair-usage limit counts crypto exchanges, each payment is a taxable disposal, and the September 2026 data breach exposed identity and transaction data. Solid for banking, honest about the crypto side-feature trade-offs.
The most aggressive rewards among exchange cards: 2-10% USDT cashback tiers, auto-interest on the card balance and 100% rebates on popular subscriptions - but it requires full exchange KYC and is explicitly unavailable in the EEA, and the platform carries the memory of the 2025 mega-hack it survived.
The EEA successor of the Bybit Card: clean 0.5% FX and uncapped 1% EUR cashback, but a separate bybit.eu account is required and the old bybit.com EEA card is closed.
Watch for the December 2026 sunset: the consumer product is closing. Solid self-custody architecture (card only moves whitelisted tokens from your Safe to settlement, up to a daily limit), but as a spending card it has an end date. Migrate to a partner app (Rebind/Picnic) or another self-custody card before 20 December 2026.
The only true bank in this comparison: a Gibraltar credit institution with deposit-guaranteed USD, savings interest paid daily in BTC (USD Savings 4.1% APY since 22 Sep 2026), worldwide metal card spending with no FX fees and high limits. But the general card cashback ended in July 2026 (now a taxi-only 3% campaign for Metal members), the $1,000/year membership makes it a wealth product, and crypto balances are NOT deposit-covered.
A US-registered, SOC-2-accredited exchange card with true point-of-sale crypto conversion - pick any listed coin as your spend wallet and it converts at the tap. The exchange and token ecosystem are mid-sized; the mechanism is the appeal.
A credible, well-backed self-custodial spend card with genuinely low fixed costs ($0 monthly/issuance/shipping, 0% APR) and true on-chain collateral custody in smart contracts. But it is not a privacy product (full KYC with SSN/national ID, GLBA data sharing), the rewards program is opaque (earn rates only in-app, statement-credit redemption at 70% of face), international FX is 'up to 3%' with a 1%-vs-2% disclosure conflict on the US side, and spending limits can be cut at Rain's sole discretion - with liquidation possible 'without prior notice'. Best for AVAX/stablecoin holders in the US (partial) and Latin America/Caribbean who want to spend against their stack; not for no-KYC users, and international terms (1-day payment window) behave much harsher than the US 21-day credit cycle.
The most complete dual-mode card: one tap switches between spending your crypto (Debit, with daily interest) and borrowing against it (Credit, from 1.9%) - with free ATM allowances up to EUR 2,000/month and no card fees. The catch: meaningful cashback needs a $5,000+ portfolio and is denominated in NEXO, and it is EEA/UK-only.
The safest pair of hands in the comparison: NASDAQ-listed issuer context, FDIC-member bank card, zero fees on USD/USDC spending and no staking requirement. The trade-offs: US-only (no Hawaii), rewards rates are variable in-app only, and non-USDC crypto spends trigger auto-sell with spread and taxes.
The most generous crypto-rewards credit card in the US market - up to 6% back in BTC or CRO with no annual fee - but the best rates require CRO staking or a Level Up subscription, BTC earn is capped by tier, and it is credit (with bureau approval), not a debit card you can top up with crypto.
A genuinely different architecture: keep full self-custody of 250+ assets and spend through a credit line instead of selling - with tiered memberships, travel cashback and premium Visa perks. The risks are equally clear: borrowing costs, portfolio volatility/liquidation risk, and a protection program that was still 'not yet live' at review time.
A Ledger-branded Visa prepaid card by Baanx that lets users spend from Ledger self-custody, converting crypto to fiat at the moment of purchase and paying 1% cashback in BTC (2% in BXX). Live across UK/EEA plus a 2025 US expansion, but the open jurisdiction list is login-gated and the old Ledger shop pages are dead links.
The most mature crypto prepaid card in the US - FDIC-member bank issued, $25k/month top-ups and clean tier structure - but the free tier earns nothing, real cashback requires a subscription or five-to-six-figure CRO stakes, and rewards are paid in volatile CRO.
The strongest physical no-KYC card pitch: a real Mastercard with ATM access, 0% FX and 0% ATM fees, XMR deposits and privacy-preserving support channels. The trade-offs are a $350 entry price, 3.5-4% deposit fees and an unlicensed custodial balance - treat it as a privacy tool, not a bank account.
The most regulated issuer context for a crypto-rewards credit card (NYDFS trust, WebBank-issued Mastercard World Elite): 4% gas/EV/transit, 3% dining, 2% groceries, 1% elsewhere, no annual fee, rewards in BTC or 50+ cryptos. It is a credit product (bureau approval, no crypto funding), US-only, and the issuer carries a 2025-2026 withdrawal-security-hold complaint pattern at the exchange level.
Gate Card exists, is actively developed and advertises the most aggressive cashback headline on this list (up to 8%), but the program is a black box: issuer unnamed, no published fees, no documented limits, regions defined only as 'non-restricted countries'. Fine as an exchange perk for heavy Gate users - hard to verify for anyone comparing card terms.
The most token-aligned card on the market: Plasma One is a real, working stablecoin neobank (instant virtual Visa, self-custodial on-chain balance, broad fiat rails, transparent fee tables, genuinely strong headline cashback) built by the XPL blockchain team - and every part of its economics points back at the token. Cashback is paid in XPL, the paid tiers are bought with 12-month XPL locks, and growth is driven by referral codes rather than organic demand. The free Lite tier is honest value (2% on the first $500/mo, free virtual card) if you treat the XPL payout as a hot potato, but the Core/Platinum tiers ask you to lock a token that has lost ~95% from its ATH for 12 months to unlock rates that then step down to 0.25-1% anyway. Use Lite, cash out weekly, and skip the locks unless you would buy XPL anyway.
The veteran (2014) reinvented as a stablecoin bank: fee-free FX worldwide, multi-currency accounts with 1:1 stablecoin conversion, real Visa/Mastercard principal memberships and an 8% headline cashback. Fee details hide in the help hub, and the El Salvador licensing move makes the regulatory story newer than its brand.
A documented, fully-priced stablecoin Visa program with named regulatory registrations - but unresolved mega-litigation changes the risk picture: Binance sued RedotPay and its founders for ~$472.8M in August 2026 over alleged user-asset diversion. Custodial balances, freeze powers, no named issuer. Treat as usable-but-contested; keep balances minimal and watch the litigation.
Back and verifiable: the Uphold Debit Card relaunched in the US with up to 4% back in XRP (Elite, $99.99/yr) or 2% (Essential, free), Visa acceptance with Apple/Google Pay, no credit check and FDIC pass-through on the USD balance. The trade-offs: rewards are XRP-denominated with conditions (holding periods, monthly caps), the program's history is pause-and-restart, and support tickets show crediting/adjustment friction. US-only (no NY/LA/territories).
A non-custodial stablecoin wallet with a Wirex-issued prepaid card: documented 1-8% cashback tiers in stablecoins, free virtual card, Apple Pay/Google Pay and up to 5% APY. The rewards are real but tiered behind $COCA balances, the official pages contradict each other on cashback rate and card network, and the independent FX spread measurement undercuts the 0% FX marketing. Verify the current card terms in-app before loading.
The original crypto-rewards card (since 2015), now a subscription product paying 3-9% in its own PLU/PLUS tokens with merchant-funded perks - and no Bitcoin option. Card spending is paused during an open-ended banking migration with a terminated Modulr contract and a claim process for stuck funds, hence watch.
30 programs ship physical cards, and every one of them charges for the format somewhere outside the headline fee table - the pattern that defines this category is that plastic's costs are lifecycle costs, not fee-table costs, which is exactly why they surprise people who compare cards on the published rows alone. The visible rows are competitive - 25 with no monthly fee, median load fee ~0% - but the lifecycle costs are where plastic pays: replacement around $100 [1], expiry every ~3 years, issuance pauses (Plasbit, August 2026 [4]), and the shipping latency on day one. This page prices the format honestly: what it costs, what it uniquely buys, and when the virtual alternative is strictly better.
The ATM is the physical card's monopoly - the one rail a virtual card cannot reach at all (wallet-NFC exceptions aside, and those exceptions cost a documented premium on the programs that offer them). Two fee layers apply: the issuer's (0-2% across this set, with monthly free allowances at Nexo, Xapo, Wirex and Vivid) and the operator's surcharge, which no program can waive. The documented ranges make the comparison concrete: Goblin's plastic at 0% issuer fee [1], Fold's at $3.50 flat or free with Fold+ [8], Xapo at 2% after $100/month with a $2,000 daily limit. The binding constraint at the machine is usually the cap, not the percentage - $250-per-24-hours programs are slow emergency rails regardless of their fee row.
| Card | ATM | Load | FX | Monthly | KYC | Regions |
|---|---|---|---|---|---|---|
| KAST | 2% + $NaN | 0% | 2% | $0/mo | Full KYC | GLOBAL, US, EEA |
| Bitpanda Card | 2% + $NaN | 0% | 0% | $0/mo | Full KYC | EEA |
| Krak Card (Kraken) | 0% + $NaN | 0% | - | $0/mo | Full KYC | UK, EEA, US |
| Fold Bitcoin Credit Card | $NaN | 0% | 0% | $0/mo | Full KYC | US |
| Fold Bitcoin Debit Card | 0% + $NaN | 0% | - | $0/mo | Full KYC | US |
| Cash App Card | 0% + $NaN | 0% | 3% | $0/mo | Full KYC | US, UK |
| Revolut | 2% + $NaN | 0% | 1% | $0/mo | Full KYC | EEA, UK, US |
| Bybit Card | $NaN | 0.9% | - | $0/mo | Full KYC | APAC, MENA, LATAM, GLOBAL |
| Bybit EU Card | 2% | - | 0.5% | $0/mo | Full KYC | EEA |
| Gnosis Pay | 2% + $NaN | 0% | 0% | $0/mo | Full KYC | EEA, UK, GLOBAL |
| Bitget Card | - | - | - | $0/mo | Full KYC | APAC, EEA, LATAM, UK +1 |
| Xapo Bank | 2% + $NaN | 0% | 0% | $83.33/mo | Full KYC | GLOBAL |
| CoinZoom ZoomMe Card | 0% + $NaN | 0% | - | $0/mo | Full KYC | US, APAC, GLOBAL |
| Avalanche Card | - | - | 1% | $0/mo | Full KYC | AG, AR, BB, BO +27 |
| KuCard | - | - | - | - | Full KYC | APAC |
| Nexo Card | 2% + $NaN | 0% | 0.2% | $0/mo | Full KYC | EEA, UK |
| Coinbase Card | 0% + $NaN | 0% | - | $0/mo | Full KYC | US |
| Crypto.com Visa Signature Credit Card | 5% + $NaN | 0% | 3% | $0/mo | Full KYC | US |
| ether.fi Cash | $NaN | 0% | 1% | $0/mo | Full KYC | US, UK, EEA, APAC +1 |
| Ledger CL Card | - | - | - | - | Full KYC | UK, EEA, US |
| Crypto.com Prepaid Visa Card | 2% + $NaN | 0% | 3% | $0/mo | Full KYC | EEA, UK, US, APAC +2 |
| Goblin Card | 0% + $NaN | 3.5% | 0% | $0/mo | No KYC | GLOBAL, LATAM |
| Gemini Credit Card | 3% + $NaN | 0% | 0% | $0/mo | Full KYC | US |
| Gate Card | $NaN | - | - | - | Full KYC | - |
| Plasma One | 2% + $NaN | - | 2% | $0/mo | Full KYC | GLOBAL, US |
| Wirex | - | 0% | 0% | $0/mo | Full KYC | GLOBAL, EEA, UK, APAC +1 |
| RedotPay | 2% + $NaN | - | 1.2% | $0/mo | Full KYC | APAC, GLOBAL |
| Uphold Card | 0% + $NaN | 0% | 1.5% | $0/mo | Full KYC | US |
| COCA | 2% + $NaN | 0% | 0% | $0/mo | Full KYC | GLOBAL |
| Plutus | - | - | - | - | - | - |
Live values from our database - details and caveats in each review. "-" = not published by the operator.
Plastic economics live in four events, not in the fee table: the order (region check, KYC, issuance fee or free [2]), the ship (~7 days tracked [1]), the live period (~3 years), and the replace (~$100 [1], plus the pause risk documented at Plasbit [4]). A metal premium is an aesthetic line item, not an economics one - Plasbit's pause covered metal first [4], and no program's metal tier has ever paid back its premium in fees. 30 of the population ships plastic; 73% also provision wallets, so most programs let the format decision stay reversible.
Warning: An issuance pause is a program-economics signal. Plasbit's August 2026 language - "overwhelming demand", waiting list - reads as supply strain, and the virtual card kept working [4]. The response playbook: keep the virtual form active, do not load balances because the plastic looks premium, and re-check the program's terms before the next load. The virtual hub covers the BIN-sponsor risk that makes pauses worth watching.
30 of the programs here, 73% with wallet support, and the ATM row is the one column where physical programs differentiate hard. Three rules survive every fee table on this page. First, know your network: in-network machines (Allpoint on Revolut US, the Sutton-Bank network behind Fold) dodge the operator surcharge that out-of-network machines add [8]. Second, size to the cap: withdrawals that exceed the daily cap are impossible at any fee, so the cap row is the emergency-planning number. Third, decline DCC at the machine - the zero-FX analysis documents why the terminal's home-currency offer is the most expensive layer in the stack (3-10%), and the ATM is where that screen bites hardest. A fourth rule is regional: the operator-surcharge layer differs by country, so a card with a named in-network partner in your country beats a nominally cheaper 0% card whose network you have never heard of - check the program's ATM-network page, not just its fee table.
13% of this page's population is self-custody - the physical format is orthogonal to the custody question, and worth stating: plastic from a self-custody program gives you the ATM rail and your own vault in one product, while plastic from a custodial program re-creates the issuer float that the self-custody hub exists to avoid. The format does not change whose balance it is; the architecture does.
73% of the fee data here is independently verified, 21 of the 30 programs carry rewards, 7 are credit products with no ATM path at all (Fold's Celtic Bank plastic [4,8]), and the median load fee is ~0% - which is to say, the fee table barely differentiates this category, and the lifecycle plus ATM economics are where the real comparison happens. Cash-first users: Goblin or Xapo by region, with the lifecycle costs priced in [1]. EEA premium-feel users: Bitpanda's published table [2]. US bitcoin earners: the Fold stack, where the plastic earns through the debit rail and Fold+ removes the ATM line [8]. And the honest closer: if your spending is 95% online and wallet-tap, the plastic premium never pays back - the virtual hub is your page, and the physical card is an emergency rail you order before you need it. The format decision, like every decision on this page, is a usage-shape decision first and a fee comparison second.
The lifecycle. Shipping (about 7 days tracked, up to 14, at Goblin [1]), replacement fees (about $100 per card at Goblin [1]), expiry every ~3 years, and issuance pauses - Plasbit suspended new plastic and metal issuance in August 2026 with a waiting list [3]. None of these appear in the load/FX/monthly rows, and together they are the plastic premium. A physical card pays for itself only if the ATM and terminal access earns it back.
Two layers: the issuer's fee and the ATM operator's surcharge. The issuer layer clusters at 0-2% - Goblin 0% [1], Nexo free to EUR 200/month by loyalty tier then 2%, Xapo free to $100/month then 2% with a $2,000 daily limit, Fold $3.50 flat (free with Fold+) [8]. The operator layer applies regardless of the program: the fee tables that say "third-party ATM operator charges may still apply" mean exactly that. Budget the operator fee separately; it is not the program's to waive.
Usually worse, in three stacked ways: the issuer's ATM fee, the operator's local surcharge, and - on non-USD programs - the FX layer from the [zero-FX analysis](/best/zero-fx/). Some programs waive their own ATM fee worldwide (Uphold Elite, CoinZoom) while operator surcharges remain. The only defense at the machine: check the fee screen, decline DCC, and prefer in-network machines where the program names a network (Fold's Sutton-Bank network, Allpoint on Revolut US).
{{stat: applepay}} of the programs here support wallet provisioning [1] - and physical cards that provision into Apple/Google Pay give you both rails from one account. The format choice is not either/or on programs with `both` form factors: plastic for the terminal, wallet for the phone. The virtual-only trade-offs are covered in the [virtual hub](/best/virtual/); the physical card's unique value is the rail that wallets cannot do.
Because terminals still decline wallet taps and some merchant categories still run imprint or deposit models - and because cash is the fallback layer when networks or programs fail. A physical card is the only format that works at every ATM on its network, which makes it the emergency rail of a crypto spending setup even for people who rarely use it. The [self-custody hub](/best/self-custody/) crowd keeps one for exactly this reason.
Daily and monthly caps, which are usually tighter than spend limits. Documented examples: Xapo $2,000/day ATM; Crypto.com prepaid daily caps $500-$2,000 by tier; KAST max $250 per withdrawal, three per 24 hours. The cap row matters more at the ATM than anywhere else - a 0% fee with a $250/day cap is a slow emergency rail. The card pages document each program's numbers with check dates.
As economics, no - metal cards carry higher issuance or replacement premiums and weight in the pocket as the benefit. As durability, marginally: metal survives wallets better. Plasbit's pause covered metal too [3], which is the reminder that premium formats are the first to stop in supply crunches. The [fee calculator](/tools/fee-calculator/) will show the metal premium never paying back in fees; it is an aesthetic purchase - price it as one, alongside the golf-club membership and the fountain pen.
Documented programs: Goblin ships internationally tracked in about 7 days, up to 14 [1]; RedotPay's terms have cited ~7 days worldwide. Shipping is also where region reality bites - a program that ships "worldwide" may still exclude your country on its restricted list, and customs can add days. Order before you need the card; the ~7-14 day window is the product's latency cost, and the virtual card in the same account is the bridge for those weeks - which is why both-format programs are the pragmatic default here.
Existing cards keep working; replacements and new formats wait. Plasbit's August 2026 pause is the current case [4]. The playbook: keep the virtual form active (most programs issue both from one account), do not load for the plastic's sake, and read the pause's language - "overwhelming demand" with a waiting list is a supply signal, not a solvency signal, but both warrant smaller balances until it resolves.
{{stat: cashback_n}} of the programs here carry ongoing rewards, and the physical format does not change the rate - rewards attach to spending, not to the card object. What the plastic does change is where you can spend: ATMs rarely earn rewards (and often trigger the ATM fee row instead), while wallet taps earn identically to the virtual card. The [cashback hub](/best/cashback/) holds the structure analysis; format-neutral.
The one whose ATM row and region list match your actual geography. If EEA: Bitpanda's published table (free issuance, 2% or EUR 2 minimum ATM [2]) is the cleanest starting point. If global-cash-focused: Goblin's 0% ATM with the $100 replacement cost understood [1]. If US: the Fold stack, where Fold+ membership removes the $3.50 ATM fee [8]. Verify the region list on the card page before ordering - it is the one check that prevents the most common physical-card disappointment.
