One multichain wallet,
or one per chain?

One seed, several key families.

A multichain wallet is not one key used on many networks. It is one seed, from which each network derives its own keys down its own standard path. Getting this straight is what makes the rest of the decision obvious.

Your recovery phrase is a BIP39 mnemonic. It produces a seed, and each chain takes that seed down a different derivation path to arrive at a different private key and a different address. In Orbit that produces four distinct address families:

• UTXO chains — Bitcoin, Litecoin, Dogecoin and Bitcoin Cash. Separate paths, separate addresses, no relationship between them beyond the shared seed.
• EVM chains — Ethereum, Base, Arbitrum and Polygon. These share m/44'/60'/0'/0/0, so they share one address across all four networks.
• Solana — derived from the seed's first 32 bytes, its own address format.
• Cardano, Tron, XRP Ledger and Stellar — each with its own scheme; in Orbit these are tracked rather than generated today.

So "one wallet for everything" is precise about the backup and loose about the keys. You write down one phrase. Behind it sit a dozen unrelated keypairs that happen to share an origin.

One thing to back up.

The argument for a multichain wallet is not that it is clever. It is that backups are where people actually lose money, and one backup is easier to get right than five.

A recovery phrase has to survive years, house moves, hardware failure and your own memory. Every additional phrase multiplies the chance that one of them is written down wrong, stored somewhere you later forget, or quietly lost. A single phrase means one procedure to do properly, once, and one thing to verify you can still read.

The secondary benefit is that you can see everything at once. Balances spread across five apps are balances you do not really know the size of.

One thing to lose.

The same property that makes the backup easy makes the failure total. This is the part a product page will not lead with, so it leads here.

If a single phrase covers thirteen networks, then anyone who obtains it takes all thirteen, in one action, with no partial outcome. There is no per-chain password, no second factor behind it, and — because this is self-custody — nobody to appeal to afterwards. The blast radius is the whole wallet.

That risk is worth accepting for most personal holdings and worth refusing for some. The dividing line is usually not the amount; it is how often the phrase gets handled. A phrase you type into a device weekly is exposed to far more than one sealed in a drawer since the day you wrote it.

A valid address is not a correct network.

If you take one thing from this page, take this one — it is specific to multichain setups and it is the most common expensive error in them.

Because Ethereum, Base, Arbitrum and Polygon share a derivation path, they share an address. Your Ethereum address is your Base address. So when you paste it somewhere, every validity check passes on every one of those networks, and the interface tells you the address is fine — because it is. What no address check can tell you is whether the sender is on the network you expected.

Assets sent to your correct address on the wrong EVM network are not lost in the sense of being gone from the chain, but recovering them is a technical exercise many people cannot complete, and for some tokens it is not possible at all. Two habits prevent it entirely:

• Confirm the network before the address. The address will look right either way; the network is the field that actually varies.
• Hold the gas token of the network you intend to use. Funds sitting on a chain where you have no gas cannot be moved off it until you get some there.

None of this is a reason to avoid a multichain wallet. It is a reason to understand what it is doing, which a wallet that shows you the derivation paths makes easier than one that shows you a logo wall.

Split by purpose, not by chain.

The usual framing — one wallet or many — is the wrong axis. Splitting by chain gives you more phrases and no more safety, because the risk was never that Bitcoin and Ethereum were in the same app.

Splitting by purpose does buy something real:

• A holding phrase. Generated once, written down, and handled as rarely as possible. Multichain is ideal here — one backup, seldom touched.
• A spending phrase. A separate wallet holding only what you are willing to have on a device you use daily. Also multichain, and deliberately small.
• Shared or business funds. Not a single-signature wallet at all. Money that belongs to more than one person needs more than one signature — that is a treasury wallet, and it is a different tool.

That is two phrases and one vault rather than thirteen wallets, and it separates the thing that actually differs: how often the key gets used.

Both, and honest about the gaps.

Orbit generates a fresh wallet on 8 networks and tracks an address you already control on any of the 13 — so a phrase generated here and a hardware wallet you keep entirely separate can appear in the same view without Orbit holding keys for either.

Keys are generated on your own device and encrypted there with a password only you know. There is no account, no email and no identity check, which also means no support desk can restore anything for you. The security model sets out what that costs.

What Orbit is not yet: a finished place to spend from. Sending from the web app is gated and bridging is not something the wallet does at all. That is stated the same way on the self-custody page and it is not going to be softened here. If you are choosing a wallet to transact from daily, that gap is the thing to weigh.

The full network table — which chains you can create on, which you can only track — is on the multichain wallet page, and it is generated from the app's own supported-chain list rather than written by hand.

Straight answers.