Coinbase status: access issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Coinbase reports received over the last 24 hours by time of day. When the number of reports exceeds the baseline, represented by the red line, an outage is determined.

Most Reported Problems

The following are the most recent problems reported by Coinbase users through our website.

• Transactions (33%)
• Website (17%)
• Mobile App (17%)
• Login (17%)
• Withdrawals (17%)

Live Outage Map

The most recent Coinbase outage reports came from the following cities:

City	Problem Type	Report Time
Paris	Withdrawals	4 days ago
Le Taillan-Médoc	Transactions	8 days ago
Leipzig	Transactions	1 month ago
Maquoketa	Website	2 months ago
West Liberty	Login	2 months ago
Houston	Mobile App	3 months ago

Full Outage Map

Community Discussion

Tips? Frustrations? Share them here. Useful comments include a description of the problem, city and postal code.

Beware of "support numbers" or "recovery" accounts that might be posted below. Make sure to report and downvote those comments. Avoid posting your personal information.

Coinbase Issues Reports

Latest outage, problems and issue reports in social media:

• 
  NEXORA (@NEXORAResearch) reported 1 minute ago

  🚨 Coinbase CEO Brian Armstrong just challenged one of the biggest trends in tech. While many companies are rushing to rebrand around AI, Armstrong says they're solving the wrong problem. His argument: AI doesn't replace digital assets, it needs them. Autonomous AI agents can't open bank accounts, can't wait days for wire transfers, and don't belong to any single jurisdiction. They need programmable, real time money, moving at a scale Armstrong says will eventually outpace all human transactions combined. That's why Coinbase is building payment infrastructure for AI agents through x402, Base and USDC, instead of just renaming the company and calling it a pivot. The companies building the actual rails may ultimately outperform the ones simply rebranding as AI.

• 
  pareidolia (@cloudsfables) reported 2 minutes ago

  @brian_armstrong Coinbase went from refusing to list anything other than BTC, even listing ETH with some skepticism, to spamming random garbage tokens to users, lol

• 
  ManLy (@ManLyNFT) reported 2 minutes ago

  CEX spot volume is down 74% since August 2025. At first glance, it looks like everyone simply left crypto. I don’t think that’s the full story. A lot of retail really did stop trading. After months of weak altcoins, repeated liquidations and no clear trend, clicking buy stopped feeling worth it. People either moved into stables, became long-term holders or left the market entirely. The data reflects that exhaustion. Crypto market cap fell 12.6% in Q2, stablecoin supply contracted for the first time since 2023 and spot volume across the top CEXs dropped another 27.9% QoQ. But the rest of the activity didn’t simply disappear. It scattered. The fastest opportunities now appear onchain first. New tokens, points, airdrops and memecoins are usually traded long before a major exchange lists them. DEX spot share is already around 15%, almost double where it was at the beginning of 2024. Even Coinbase reported that DEX volume inside its own app doubled QoQ. Active traders also moved toward perps, prediction markets and tokenized assets because holding spot through a slow bleed offers very little excitement. Prediction market volume grew nearly 49% in Q2 while CEX spot activity continued to shrink. Institutions have also changed how they enter the market. They can now access crypto through ETFs, traditional brokers and CME products without opening another exchange account or managing wallets. CME crypto trading activity was up 44% YoY in the first half of 2026. Then there is the trust problem. After years of exchange collapses, hacks, frozen withdrawals, KYC friction and regional restrictions, many users no longer want their entire portfolio sitting behind one company’s login. CEXs are still essential for fiat access, deep liquidity and onboarding. But they are slowly becoming a bridge rather than the final destination. That is why exchanges are adding DEX routing, prediction markets, payments, tokenized assets and almost every product they can fit into one app. So this chart is not only showing a quiet market. It is showing a completely different market structure. The next retail wave probably won’t return to one exchange and trade the same spot pairs all day. It will be split across CEXs, wallets, DEXs, perps, ETFs, prediction markets and tokenized assets. Volume didn’t just disappear. The spot tab lost its monopoly.

• 
  Shadow/ (@cryptowildwild) reported 4 minutes ago

  @SPCMNandHOBBES A dev tax, a foundation, a sketchy launch/history, transparent chain with fungibility issues, exchange boot lickers, exploited privacy, and a community that loves coinbase and etfs

• 
  Donald S Pritt Jr (@PrittJr) reported 5 minutes ago

  @shawmakesmagic @brian_armstrong Yeah @coinbase is terrible. They forced sold my crypto without consent. I will be posting all over social media .

• 
  Charu (@Charu_Sethi) reported 8 minutes ago

  The Model Context Protocol locks its authorization spec on Monday. It's the default way AI agents connect to tools and data, and the big change is auth: it now uses the same OAuth building blocks a bank's API gateway would. So an agent can prove which server it's talking to, and which token is good for which resource. Here's the part that matters. That secures the channel. It doesn't secure the wallet. An agent proving it's allowed to call a payments API still has no standard, revocable, auditable limit on how much of a treasury it can actually move, tied to a named person or company who authorized it. Settlement is handled. Agents can pay on-chain today. Identity-to-server is getting handled now, with MCP. The bit in the middle, the spend mandate, is still the gap. The on-chain candidates for it, ERC-8004 and ERC-8273, are both still Draft. My read: the agent stack is maturing from the outside in. The plumbing on either end lands first, and the hard middle, bounded spending authority a compliance team can defend, lands last. For people building agent payment infra: is the mandate layer something you're solving in the wallet today, or waiting on a standard for? MCP is an open spec, no token; this is a publication milestone, and the maintainers say nothing switches off on the date. @AnthropicAI @coinbase @ethereum #AIagents #AgenticPayments

• 
  Pharos (@PharosInsights) reported 9 minutes ago

  The strategic prize may be an agentic financial control plane: one mandate and one risk policy that can route a transaction across an internal ledger, card, bank account or stablecoin—and stop, explain or repair it when necessary. Coinbase can still win that market. Its strongest AiFi thesis is not that every agent payment must settle on Base. It is that Coinbase can combine USDC, Base, x402, wallets, security, compliance, exchange liquidity and developer distribution into the orchestration layer. That also changes how x402 should be measured. One human instruction may generate thousands of API calls, data purchases and compute events. But usage events, payment instructions and final settlements are three different metrics. Efficient systems will also batch, net and pre-fund many machine transactions. More machine activity does not automatically mean one blockchain settlement per action. A July measurement study found 136.7 million Base x402 settlements worth $44.1 million—about $0.32 each on average. It classified 21.2% as fictitious and 63.8% as internal to linked clusters. Internal does not automatically mean fake, and low value is expected in micropayments. But the conclusion remains: Transaction count is not adoption. The relevant KPIs are independent buyers, external revenue, repeat usage, verified work and economically necessary settlement. The largest missing layer is accountability. A blockchain can prove that a key signed. It cannot prove that the agent understood the user’s intent, purchased the correct service, received a valid result or deserves credit. A wallet gives software purchasing power. It does not give it judgment, legal identity or a balance sheet. ✅

• 
  Kelsey Cody (@Harpendenblinds) reported 10 minutes ago

  Hi! Sorry you’re experiencing this. If you staked through Coinbase Wallet, make sure you’re using the same wallet and recovery phrase. Your funds are typically still on-chain even after the app update. Let me know if you need guidance to fix this.

• 
  Igorek (@ipushyl) reported 12 minutes ago

  @WatcherGuru Damn woke up and saw this wtf had to emergency transfer assets to Coinbase

• 
  dnsn (@imdnsn) reported 12 minutes ago

  @brian_armstrong gm Brian, i have a problem on Coinbase and support isnt helping so i figured id go straight to the big guy. Someone created an account with my stolen info and tried to trade with funds that werent settled. I had an account and now im trying to use it again but im perm banned. hlp

• 
  Donald S Pritt Jr (@PrittJr) reported 17 minutes ago

  @_markjones1 @coinbase @brian_armstrong Help ??? They already sold my crypto without my consent. That’s brazen and harmless to people using the platform.

• 
  Will Hodl (@WillHodlMusic) reported 23 minutes ago

  @MicroSeed_io When you hold your coins on Coinbase, you're not the customer, you're the product.

• 
  PT (@spartywrx) reported 26 minutes ago

  @jbmillen @BitPaine When the holder wants to sell using Coinbase those coins are flagged and seized at the CEX level Many ways around that issue but that’s a main point

• 
  wendysli (@marver_wen62435) reported 27 minutes ago

  @Coachjv_ Is there anyone out there that can help me? I just got scam 25,000 of bitcoin the Coinbase and they say there’s nothing they can do. Does anyone have any advice for me please and thank you

• 
  marie_andrea (@crypto_andream) reported 28 minutes ago

  @XMoney @premium Can you please fix the issue with adding a Coinbase card to X money. 🫶🏼

• 
  JesTer396 🏴‍☠️ (@JesTer396) reported 30 minutes ago

  Nobody needs this, how about you start with customer service and then lets talk, Coinbase failed 1st grade and now this guy want to fill the digital space with bots.

• 
  ferk.inj (@ferkoss78) reported 31 minutes ago

  @injective @coinbase Yet, price is going down and near all time low...something really missed or not seriously taking care to by the injective team. Clearly your achievements are definitely not enough on the price side.

• 
  G@G@ (@cryptiogaga) reported 32 minutes ago

  9/ Coinbase Adds Support for BRC-20 Tokens Coinbase wallet now supports Bitcoin-based BRC-20 assets, enabling transfers and swaps for ORDI, SATS, and more.

• 
  rob | skydrop.sol (@sneakbots) reported 34 minutes ago

  @shawmakesmagic @brian_armstrong Send it on chain, swap to usdc, send it back to coinbase. Dumb as **** but you need to do this to avoid getting bent over

• 
  Lea Thompson (@LeaT_Design) reported 35 minutes ago

  @GoingParabolic @cobie cobie ain't your personal crypto police, buddy. coinbase support exists for a reason. good luck tho lol

• 
  Holden Culotta (@Holden_Culotta) reported 38 minutes ago

  “Palantir and Polymarket are partners.” “The goal is not just … mass surveillance.” “It’s predictive policing on a massive scale.” “This is the privatized panopticon.” Whitney Webb just exposed the truth about Polymarket’s origins. And it’s much darker than you think. The “official narrative” is this: Polymarket founder and CEO Shayne Coplan created the platform during Covid, because that time of uncertainty was “the best time for prediction markets.” But that narrative, Webb says, is “bunk.” “He had all these connections, particularly the people that bridged the US-Israeli crypto tech scene.” And the idea for Polymarket really came from DARPA, the US government’s Defense Advanced Research Projects Agency. “There’s an interview that he gave in 2020 where he said a lot of the passion for the creation of Polymarket … was DARPA’s Policy Analysis Market.” “Even though he’s alone in his bathroom building this company … he somehow raises $4 million just three months after the company launches.” “The main people backing him are Coinbase’s earliest employees.” “Remember, Coinbase takes Augur and sunsets it.” Augur, another prediction market platform, was an “attempt to resurrect PAM.” “What was PAM?” “PAM, or the Policy Analysis Market … was part of TIA.” “Total Information Awareness.” TIA was another DARPA program launched after 9/11 to develop a government mass surveillance apparatus. Congress defunded it after public scrutiny and backlash. But TIA never went away. It was privatized and eventually turned into Palantir. Webb: “The goal was not just to see everything going on in the present … in a sea of data, of everyone’s data in mass surveillance.” “It was to predict.” “Pre-crime.” “Polymarket … just like with PAM, the odds are there to help feed the pre-crime engine.” “We know that Polymarket bets can be rigged.” “And that Polymarket itself puts out misinformation sometimes to make trades swing one way or the other.” “And that insider trading is a huge problem.” “The context of that, as it’s meant to interface with what is now Palantir, is really unsettling.” Mark Goodwin: “And not only privatized, they’re … making money off it.” “Insider trading is not a problem, it’s a feature.” @_whitneywebb @markgoodw_in

• 
  Jeff Park (@dgt10011) reported 41 minutes ago

  Big news today that CME launched single stock futures - these will trade 23hrs/day, weekends, cash-settled leverage, on the 55 most liquid equities names in the US market! But this is actually not the first time CME has tried to do this. In fact, they made a huge effort in 2002- and failed fantastically. It's worth studying what happened, what's different now, and what this means for the frontier of finance. Time for a side quest- That story starts with the Commodity Futures Modernization Act of 2000. This was the ugly byproduct of a 20yr reg turf war between the SEC and CFTC (sound familiar guys?), in which margin rules/short sales/reporting requirements were so complicated that even after approval broker dealer compliance framworks had no idea how to handle it. Most people don't remember this brief glitch in history because this was in fact just a big sideshow for what was actually the bigger/$$$ gamble of a darker legacy - exempting OTC derivatives from regulation altogether; this would in turn allow the CDS market to expand without oversight, and we all know what happened after that... But perhaps more practically speaking in fairness to the regulators, the more obvious reason why SS futures failed at the time was probably that there was just no genuine economic advantage over existing tools- because you can in fact get most of the leverage via listed options! Maybe you got slightly better capital efficiency for directional bets, but it wasn't good enough to otherwise offset the biggest reasons why most financial products fail: fragmented liquidity. So why are they attempting to do this again? There are many factors you can point to, but the single biggest difference now is that the retail market structure has changed enormously. American retail is more comfortable with leverage than ever in history, zero commission trading has brought in surplus liquidity beyond wildest dreams, levered ETFs and 0DTE options cant grow fast enough, and crypto perps have demonstrated that there is a ton of appetite for this kind of risk that simply just didn't exist before 2008. If you look at what the CME has been doing, the "retailification of leverage" has been happening for a long time already, starting with micro bitcoin futures called "BFFs" (Bitcoin Friday Futures lol) and the cringy Gen Z social ad that followed for those that are still scarred from seeing it. It's why CME launched for SS futures standard contracts (100 shares) but also micro contracts (10 shares). Honestly - who needs 10 shares futures contracts? The other reason, and probably more critically important, is to pursue a defensive posture. The CME doesn't compete in a vacuum anymore - with exchanges like Coinbase, Robinhood (+ a JVs between Susquehanna for predictions market), Hyperliquid all going after the same retail customer, the race to become the "everything brokerage" is just as much about distribution as much as it is about product design. And we know distribution is everything in finance because the beauty of Reg T + futures based margining is at its most competitive with scale and diversity, for the physics of money operates like a gravitational field: the larger the mass, the stronger its pull. So what does this mean for you? The single stock futures is not the product. You are the product. YOU are the yield. You are the currency that feed these HFTs, market makers, "big finance" just like you were the data currency for social networks, marketers, "big tech." Your intent to trade, the order book you create as makers, the tape you paint as takers, your inability to discriminate for best price when there are tens of seemingly fungible but slightly variant risks being offered, is their profit. And while it may all sound really dire when I frame it like this, but there is actually a way to win. And it goes back to the first principle of why the single stock futures failed to gain traction back in 2002. You must trade OPTIONS. Of course none of this is actual investment advice and you must always do your own research. But as I've written many times before on X already, options are the best tools retail investors have to protect against big finance. That's because options have the greatest asymmetric leverage embedded in the physics of its product that allow great convexity with great duration. When CME's Duffy says "perps are bad products for retail" he is not necessarily wrong- perps have the potential to be the most dangerous products for retail because they have no assurance or guarantee to control their own outcomes, especially given retail is so small- institutions can liquidate you (or each other, and you're just an ant caught in a stampede of bulls) where you have no agency. The reality is that the commodities futures market since the beginning of time has been found useful because it combines speculators WITH natural hedgers. And there is no natural hedger on earth who would take perps risk to hedge their long term business. Duration is an asset. Duration deserves a premium. Term structure exists because there is in fact a market for time. And when you own an option, it means you have the choice, but never the obligation, to meet time where and when you demand it. As I write this, I'm reminded that history has a peculiar sense of humor, delighting in the ironies of fate. The same bill that allowed then failed single stock futures market in 2000 is the same bill that gave us the reg vacuum for the CDS market that basically is the single biggest proximate legal cause of the 2008 financial crisis. And twenty years later yet again now as we head into CLARITY posturing for the next two weeks, on another epic settlement for a CFTC vs SEC battle, there are public debates occurring on various salient features that the crypto industry cares about. But you would all be wise to take note that if the past precedence holds again, the most consequential thing that will happen will actually be interpreted as a footnote, just like the "the Enron loophole" (aka. the OTC swap exemption) and it WILL involve offshore derivatives just as it did last time. And that footnote is what is going to let crypto industry expand again, bigger, stronger and faster. Because that is the other physics of money beyond a gravitational field: the harder you try to confine capital, the faster it leaks across borders.

• 
  RiskOff (@riskoff_hl) reported 45 minutes ago

  This week on @HyperliquidX the platform grew while the token got sold. RWA markets ran 54% of total volume, real-world assets outpacing crypto perps for the first sustained stretch in the platform's history. HIP-3 RWA volume hit $26B on its own. Open interest climbed to a 2026 high near $11.5B, the strongest since the October 2025 crash. By every structural measure, this was a big week. $HYPE didn't care. The token slid from $62 to $59, down about 5% for the week, with an intraweek low near $56-57. ETF flows didn't help either, $8.6M walked out the door for the second week running. Here's why: Multicoin Capital put roughly $116-120M in $HYPE into the unstaking queue starting July 22-23. Selini and Galaxy Digital added more behind them. The base overhang runs around $150M, clearing near July 28. Some estimates that include Paradigm-linked wallets push closer to $291M, though that figure is less confirmed. Once these tokens clear, they could get sold, restaked, or simply moved to new wallets. Multicoin calls the move wallet rotation, not a sale. Some of the tokens have already landed on Coinbase. Not everyone's heading for the exit. One wallet staked $32.9M on July 25, sitting on months of gains and choosing to lock in deeper instead of cashing out. Revenue came in at $7.37M on $10.65M in fees, consistent with the slower pace lately. Liquidations stayed calm, no cascades, shorts taking most of the pain. The platform is doing everything right. The token is getting sold anyway. July 28 is when we find out if that's temporary or the new normal.

• 
  Tait (@taitsgambles) reported 47 minutes ago

  @shawmakesmagic @brian_armstrong Yeah icl if you don’t have Coinbase 1 or prime idk what it’s called then you get ****** on fees I’m paying them for it…. Kinda fkn **** but yeah Tried Karaken, similar, you need their premium to actually get anything half decent (but you are again paying them)

• 
  CleevioX (@cleevio) reported 47 minutes ago

  For a couple of years, every tokenized product we watched launch had the same problem. The thing shipped fine, then sat there because there was no one on the other side of the trade. That changed this week. Tokenized-asset holders crossed 1 million, up more than 200,000 in seven days after Robinhood opened tokenized stocks to a global base. Around the same time, Mubadala Capital put a $75M private-markets fund on Base, Solana, and Sui, and Coinbase took a position on its own balance sheet. A holder base that size is what a tokenized product needs to be worth issuing. It just took longer to arrive than the tokens did.

• 
  ₿oomer ₿itcoiner Gus (@BTCGUS21) reported 49 minutes ago

  @bramk I think at the end of the day, it's simpler than that! The BIP-110 fork is an unfortunate certainty! But, I do not see anyone with more than 20 bucks worth of Bitcoin exchanging their BTC for BTZ. Back in the day, big block folks rushed to exchange their BTC for BCH. I believe Coinbase immediately listed both; other exchanges took longer, and some wallet providers never supported BCH! But the smart ones who suddenly had PKs to two ledgers quickly did the opposite, in essence doubling their bitcoin stack! Anyway, my oversimplification is that the chain that retains the BTC ticker wins!

• 
  Ordi (@ordi_brc) reported 51 minutes ago

  @ChainRacingClub Big if true, though cannot find confirmation Would love to see @coinbase support the Bitcoin ecosystem by listing spot ORDI

• 
  Rob (@RobGuerra90) reported 52 minutes ago

  Financials leads with 102 companies and 178 calls, ahead of Info Tech's 52 and 124. But the standout: Coinbase (COIN) tops the board at 10 calls while its momentum is falling — as Block, MARA and MSTR keep rising.

• 
  StreamSpot (@streamspot) reported 52 minutes ago

  @stambouli_o1 @coinbase Bro i need your help and you ignore me?

• 
  Kshitij Shah (@stuli1989) reported 57 minutes ago

  @rina_rrnaaaaa Hey Carine, just dropped a DM to Coinbase Support - let me know if you want me to drop one to you directly as well.