Trezor vs Ledger
Both keep your keys off an internet-connected computer. They differ on open source versus certified silicon, on what each has actually been caught out by, and on what their desktop apps tell the company about you.
Short answer: Trezor if you want firmware you can inspect and a company that has not shipped a key-export feature. Ledger if you want a certified secure element in every model and the widest coin support. Both are enormously better than leaving a balance on an exchange, and the argument between them is narrower than the internet suggests.
What follows is the honest version of the comparison, including the parts each company would rather lead past.
The trade-off everything else follows from
Trezor's firmware is open source. Anyone can read it, and people do. The cost, historically, is that the Model One and Model T have no secure element — the seed lives in general-purpose memory, and someone with the device in hand and the right equipment can get it out. Kraken Security Labs demonstrated exactly this in 2020, extracting seeds from both models with about fifteen minutes of physical access. Trezor's newer Safe models added a secure element, which closes that gap.
Ledger has always used a certified secure element, the same class of chip in a passport or a bank card, which makes physical extraction genuinely hard. The cost is that the code running on that chip is closed. You are trusting Ledger's word and their auditors, not your own reading.
That is the real fork in the road: verifiable code with a historically softer physical shell, or a hard shell you cannot look inside. Neither answer is stupid, and which one is right depends on whether the attacker you actually fear is holding your device or sitting in a supply chain.
What each has actually been caught out by
Reputation is worth less than incidents, and both have some.
Ledger's 2020 customer database breach exposed around a million email addresses and roughly 272,000 detailed records — names, postal addresses, phone numbers — from its e-commerce system. No keys were involved and no device was compromised. What leaked was a list of people known to own crypto, with their home addresses on it, and the phishing and the reported real-world threats that followed are the reason that breach still matters. If you are weighing this, our piece on wrench attacks is the relevant context: the practical risk to a hardware-wallet owner is increasingly physical rather than cryptographic.
Ledger Recover, announced in 2023, was an optional paid service that shards your seed across three custodians so it can be restored. The backlash was not really about the service. It was about what the service proved: the firmware could be updated to export key material from the secure element, which many owners had understood to be architecturally impossible. Ledger's position — that the device only does this with explicit consent — is true and does not undo the point that the capability exists.
Trezor's weakness has been physical, as above, plus the general property that an open device invites documented attacks. That is arguably the system working: the extraction technique is public, which is why the Safe line was built.
Privacy, which is where they are closer than you would like
A hardware wallet protects keys. It does nothing on its own about what your desktop app tells its maker.
By default, both Ledger Live and Trezor Suite fetch balances and transaction history from company-run servers. That means the operator can associate your IP address with a set of addresses and the balances behind them. For a device bought to keep custody personal, that is a meaningful leak, and it is on by default in both.
Both let you fix it. Trezor Suite ships a Tor toggle in the interface and supports pointing at your own Bitcoin node, which is the better arrangement out of the box. Ledger Live can also be configured against your own node, with more effort. If privacy rather than only custody is the reason you are buying, do this on day one — it is a five-minute change that undoes the default.
Coins
Ledger supports more assets, and it is not close. If your holdings are long-tailed, that decides it.
For Monero specifically, both work as a signing device with the Monero GUI — Ledger Nano and Trezor Model T are the supported pairings — with the desktop wallet doing the chain scanning and the device holding the spend key. Note that this protects the key and does nothing for chain privacy; which wallet you scan with is the decision that affects that, covered in best Monero wallet.
Which to buy
If you want to be able to verify what your device is running, and your realistic threat is a compromised computer rather than a burglar with lab equipment: Trezor, and a Safe model if you want the secure element too.
If you hold a wide spread of assets, or you are specifically worried about someone getting hold of the device: Ledger.
Two things matter more than the choice. Buy directly from the manufacturer — a supply-chain tampered device is the one attack neither design defends against, and marketplace listings are where those come from. And write the seed on paper, never into a phone, a photo or a password manager, because every large loss that was not a scam was a seed stored somewhere convenient.
Funding it without an account
Once the device is set up, it needs coins, and buying them on a verified exchange first means the first thing your new wallet does is receive a transfer already tied to your name. Swapping an asset you hold avoids that: BTC to XMR, ETH to BTC and the rest of the pair pages are non-custodial and need no account. The anonymous crypto wallet guide covers how the custody and privacy questions fit together.
Common questions
Is Trezor or Ledger more secure?
They fail differently rather than one being safer. Ledger's secure element makes extracting a seed from a device in an attacker's hands much harder. Trezor's open firmware means the code holding your keys can be independently audited, and the newer Safe models have a secure element as well. Against the threat most people actually face — malware on the computer they sign from — both are equally effective, which is most of the answer.
Is Ledger still safe after the Recover controversy?
No key material has been shown to leave a device without the owner's consent, and no exploit came out of it. What changed was what owners understood the device to be: the firmware can be updated to export a sharded seed, so the guarantee is Ledger's policy rather than physics. If your threat model included "even the manufacturer cannot get this out", that assumption was wrong before the announcement too.
Can a hardware wallet be hacked remotely?
Not in the way people picture. The private key never leaves the device and signing happens on it, so malware on your computer cannot read the key. What malware can do is swap the destination address you are about to approve — which is exactly why the device has a screen, and why confirming the address on that screen rather than on your monitor is the step that must not be skipped.
Do I need a hardware wallet for a small balance?
For an amount you would shrug at, a well-run software wallet on a clean device is reasonable. The threshold is not a number so much as a question: would losing this hurt? Hardware wallets solve one problem — a compromised computer — and they solve it completely.
What happens if the company disappears?
Nothing happens to your coins. Your seed phrase is the wallet; the device is a way of using it. Both Trezor and Ledger use standard derivation, so a seed from either can be restored into other wallets. Test this before you need it rather than after.
Should I buy a used hardware wallet?
No. A tampered device — pre-initialised with a seed the seller knows, or physically modified — is the one attack the design cannot defend against, and the discount is never worth it. Buy from the manufacturer directly.
