What Is zkEVM?

In a classic blockchain, every node verifies the result by re-executing all the transactions in a block. In a zkEVM, transactions are executed once, and a prover produces a short proof showing that this execution was carried out according to EVM rules. A smart contract on the Ethereum mainnet can verify this proof and accept the result of thousands of transactions without recomputing them. This lowers the cost per transaction.

Because the EVM was not designed with zero-knowledge proofs in mind, making it provable is technically difficult. For this reason, projects choose different levels of compatibility: some aim to be exactly equivalent to Ethereum and give up proof generation time, while others accept some differences from the EVM to make proof generation easier or compile Solidity code to their own virtual machines. The type classification proposed by Vitalik Buterin is widely used to describe these differences. Networks such as Scroll, Linea, ZKsync Era and Taiko are examples of different zkEVM approaches.

The advantage of zkEVMs is that developers can use existing Ethereum tools and contracts largely without modification. Their challenges are that proof generation requires intensive computing power and that the complexity of proof systems makes the risk of bugs high. The use of zkEVM-based verification on Ethereum's main layer is also being researched.