If you have used a smart contract wallet on Base, such as Coinbase Smart Wallet, you may have noticed something odd. You get an address the moment you create the wallet, and you can receive funds at it right away, yet the wallet itself has not been deployed onchain yet. No contract exists at that address until you send your first transaction.
This is not a trick. It is a deliberate feature of the Ethereum Virtual Machine called CREATE2, and understanding it helps explain why smart wallets behave differently from the simpler wallets you may be used to.
Two ways a contract can get an address
Every smart contract on Base needs an address, the same way every wallet needs one. There are two opcodes that can assign that address when a contract is deployed: CREATE and CREATE2.
With the original CREATE opcode, a new contract's address depends on the deployer's account and how many transactions that account has already sent, known as its nonce. This works fine, but it means you cannot know a contract's future address with full certainty until it is actually deployed, because the nonce keeps changing as other transactions happen.
CREATE2 was introduced through Ethereum Improvement Proposal 1014 and activated in the Constantinople upgrade in 2019. It changes the formula. Instead of depending on a changing nonce, the address depends on three fixed inputs: the deploying factory contract's address, a chosen value called a salt, and the contract's initialization code. Feed the same three inputs into the formula and you always get the same resulting address, whether the contract has been deployed yet or not.
That last part is the key. Because the address can be computed in advance, it exists as a piece of math before it exists as a deployed contract. An address that is reserved this way, but has no code at it yet, is often called a counterfactual address.
How this applies to a smart wallet
A smart contract wallet is itself a contract, not a simple key pair like older wallet types. When you set up a smart wallet, the app computes your wallet's future address using CREATE2, based on your account details and a factory contract maintained by the wallet provider. That computed address is yours from that moment on.
You can hand that address to someone and receive tokens at it, because receiving funds does not require any code to exist at an address. The Base network simply records that the address now holds a balance. Your wallet contract gets deployed later, typically bundled into your first outgoing transaction, at which point the deployment cost is paid alongside whatever else that transaction does.
This is also part of why some smart wallets can offer the same address across several EVM compatible networks. If the same factory contract, the same salt, and the same initialization code are used on each network, the CREATE2 formula produces the same address everywhere, even though each network's copy of the contract is deployed independently.
Why this design is useful
The practical benefit is a smoother setup experience. You get a usable address immediately, without needing to pay gas or send a transaction just to bring a wallet contract into existence. The deployment happens automatically the first time you actually need it, so the cost is folded into real activity instead of being an upfront tax on simply having a wallet.
It also means a wallet address can be shared and funded ahead of time, which is convenient for onboarding someone new, setting up a wallet for later use, or receiving funds before you have decided exactly how you will use the account.
What to keep in mind
CREATE2 is a well established, audited part of the EVM, not an experimental feature, and the counterfactual pattern is widely used across account abstraction wallets, not something specific to one provider. That said, a few things are worth understanding rather than worrying about.
First, an undeployed address having a balance is normal, not a sign that something has gone wrong. If you check a brand new smart wallet address on a block explorer and see no contract code, but a token balance shows in your wallet app, both of those things can be true at once.
Second, the security of a CREATE2 address depends on how the salt and owner credentials are set up by the wallet's developers. Reputable wallet providers tie the salt to your specific account so that only you can ever cause your intended contract to be deployed at that address. This is an implementation detail handled by the wallet software, not something you configure yourself, but it is why using well audited, established wallet providers matters more than it might for a simpler account type.
Third, a deployed contract can only be replaced at the same CREATE2 address in very specific, deliberately engineered circumstances, and standard wallet factories are not designed to allow arbitrary redeployment. This is a known area of technical scrutiny in smart contract security research, which is one more reason to prefer wallet contracts that have gone through public audits.
The short version
CREATE2 lets a contract's address be calculated in advance from fixed inputs, rather than assigned only at the moment of deployment. Smart contract wallets use this to give you a working, fundable address right away, while the actual wallet contract deploys later, usually with your first transaction. It is a standard, well understood piece of Ethereum infrastructure behind a detail that can otherwise look confusing the first time you notice it.