Published on Sat Aug 22 2026 00:00:00 GMT+0000 (Coordinated Universal Time) by Jacob Cavazos

Every transaction on an EVM blockchain requires gas. Gas is the fee paid to validators (or sequencers, on L2s) for including and executing your transaction in a block. On Ethereum mainnet, gas has historically been the single largest barrier to onboarding new users. On Base, gas is cheap, often fractions of a cent per transaction, but the requirement still exists: you need ETH in your wallet to pay for it.

This creates a catch-22. A user bridges USDC to Base, wants to swap it for another token, and cannot. They have no ETH. They cannot swap. They cannot move their USDC. They are stuck until they acquire ETH, which usually means bridging more assets or finding an exchange that supports Base withdrawals.

Gasless swaps solve this problem. They let users trade tokens without holding ETH for gas. The concept is simple. The implementation varies. This article explains the three general approaches that make gasless trading possible, how each works at a high level, and what tradeoffs they carry.

Why Gas Exists

On Ethereum and EVM-compatible chains like Base, every state change requires computational work. A simple token transfer updates two balances. A swap through an AMM updates reserves, transfers tokens, and may trigger callbacks. The more complex the operation, the more gas it consumes.

Gas serves two purposes. First, it compensates the network for the computational resources consumed. Second, it prevents abuse. Without gas costs, an attacker could spam the network with infinite loops and computationally expensive operations at no cost.

On Base, gas costs are dramatically lower than Ethereum mainnet. Base uses an OP Stack optimistic rollup architecture, and transactions typically cost a fraction of a cent. But the requirement to hold ETH remains. For a user who has only ever held stablecoins, this is a real friction point.

The Problem: No ETH, No Transaction

The standard EVM transaction flow is rigid. An externally owned account (EOA) signs a transaction with its private key. The transaction includes a gas payment in the native token (ETH on Base). The transaction is invalid without sufficient gas payment.

There is no native mechanism for a third party to pay gas on behalf of an EOA. No way to pay gas in ERC-20 tokens. No way to batch multiple operations into a single signature. The protocol enforces these constraints at the consensus level.

This is why gasless swaps exist as a category. They work around these constraints using one of three approaches.

Approach 1: Signature-Based Approval (Permit2)

The first approach to gasless trading does not eliminate gas entirely. It eliminates the need for the user to send multiple transactions. The key innovation is the off-chain signature.

How Traditional Token Approvals Work

Before you can swap a token on a DEX, you must approve the DEX’s smart contract to spend tokens from your wallet. This is an on-chain transaction. It costs gas. On Ethereum mainnet, this used to cost $5 to $20 per approval. On Base, it costs fractions of a cent, but it still requires ETH.

If you want to swap on three different DEXes, you need three separate approval transactions. Each one costs gas. Each one requires ETH.

How Permit2 Changes This

Permit2 is an open-source token approval contract created by Uniswap Labs. It is a public standard that any protocol can integrate — for a developer-focused explanation of how signature relay works, see Permit2 and Gasless Transactions. The core idea is simple: approve Permit2 once, then use off-chain signatures to authorize individual spends.

The flow works in two steps:

1. One-time on-chain approval. The user approves the Permit2 contract to spend their tokens. This is a single transaction that costs gas once. After this, the user never needs to approve another contract individually.

2. Off-chain signature for each spend. When the user wants to swap, they sign an off-chain message (not a transaction) that authorizes a specific contract to spend a specific amount for a specific duration. This signature is free. It costs no gas. The protocol includes the signature in the settlement transaction, and Permit2 verifies it on-chain.

Permit2 works with any ERC-20 token, regardless of whether the token supports EIP-2612 (the earlier permit standard). It adds features like expiring approvals, batched approvals, and signature-based transfers that bypass allowances entirely.

What Permit2 Does and Does Not Solve

Permit2 solves the approval problem. It does not solve the gas payment problem. The final settlement transaction still requires gas, and someone must pay it. In most Permit2 integrations, the user still sends the settlement transaction and pays gas in ETH.

What Permit2 does is reduce the number of transactions from two (approve + swap) to one (swap only, with the signature included). On Base, where gas is cheap, this is a meaningful UX improvement. But it does not make the swap truly gasless for the user.

Approach 2: Account Abstraction and Paymasters

The second approach uses account abstraction to separate gas payment from the user. This is where true gasless transactions become possible.

What Account Abstraction Is

ERC-4337 is the Ethereum standard for account abstraction without consensus-layer changes. It introduces a separate mempool for UserOperations, which are pseudo-transactions that represent user intent. A special actor called a bundler packages these UserOperations into a single transaction and submits it to the EntryPoint contract.

The key innovation is the Paymaster. A Paymaster is a smart contract that can pay for gas on behalf of the user. When a UserOperation includes a Paymaster address, the EntryPoint contract asks the Paymaster whether it is willing to cover the gas cost. If the Paymaster approves, the user pays nothing.

How Paymasters Enable Gasless Swaps

A Paymaster can implement any logic it wants. Common patterns include:

• Sponsored gas. A dApp deploys a Paymaster that covers gas for all users interacting with its contracts. The dApp pays the ETH. The user pays nothing.
• Token payment. The Paymaster accepts ERC-20 tokens (like USDC) from the user and uses its own ETH to pay gas. The user effectively pays gas in USDC instead of ETH.
• Signature-based sponsorship. The Paymaster verifies an off-chain signature from an authorized signer before agreeing to sponsor the operation.

On Base, where gas costs are sub-cent, sponsoring gas for users is economically viable for protocols that generate revenue from swaps or other operations. The cost to the protocol is negligible. The benefit to the user is significant: they never need to hold ETH.

Tradeoffs of Account Abstraction

Account abstraction adds complexity. The bundler infrastructure must be reliable. The Paymaster must be funded. If the Paymaster runs out of ETH, all sponsored operations fail. The user is trusting the Paymaster to remain solvent and available.

There is also a latency consideration. UserOperations go through a separate mempool, and bundling adds a small delay compared to direct transaction submission. On Base, this delay is typically a few seconds, which is acceptable for most use cases.

Approach 3: Intent-Based Execution

The third approach is the most recent and the most powerful. It eliminates both the gas problem and the execution problem by separating user intent from on-chain execution entirely.

What Intents Are

An intent is a signed off-chain message that specifies what the user wants, not how to achieve it. For example, a user signs a message saying “I want to receive at least 1,000 USDC for my 1 ETH.” This message is not a transaction. It cannot be executed on-chain directly. It is an expression of trading constraints.

A solver (sometimes called a filler or relayer) reads this intent, determines the optimal execution path, and submits the settlement transaction on-chain. The solver pays for gas. The solver executes the trade. The user does nothing except sign the intent.

How This Enables Gasless Swaps

Because the solver submits the transaction, the solver pays the gas. The user never interacts with the blockchain directly. They sign an off-chain message, wait for it to be filled, and receive their output tokens.

This is how UniswapX works. Users sign orders specifying their swap parameters. Fillers compete to satisfy these orders using whatever strategy they choose. The winning filler submits the settlement transaction and pays gas.

1inch Fusion works similarly. Users sign intents that are filled by resolvers through a Dutch auction mechanism. The resolvers cover gas costs.

CoW Protocol uses a batch auction model where solvers compete to find the best settlement for batches of user-signed intents. Solvers pay for execution.

What Makes Intents Different

Intent-based execution is fundamentally different from the other two approaches. Permit2 and account abstraction both modify how transactions are submitted and paid for, but the user is still interacting with on-chain liquidity directly. Intents abstract the entire execution layer.

This has several consequences:

• MEV protection. Because the user’s intent is not visible in the public mempool, MEV bots cannot sandwich it. The solver bears the MEV risk and factors it into their pricing.
• Gasless by design. The solver pays gas. The user pays nothing.
• Better pricing through competition. Multiple solvers compete to fill the same intent, which can result in better prices than executing directly through a single AMM pool.
• Surplus capture. If the solver finds a better price than the user’s minimum, the solver can keep the surplus as profit. This is the economic incentive that drives solvers to participate.

Tradeoffs of Intent-Based Execution

The user is trusting the solver infrastructure. If no solver fills the intent, the swap does not execute. In Dutch auction models (like 1inch Fusion), the price degrades over time until a solver is willing to fill it, which means the user may get a worse price if they are impatient.

There is also a custody consideration. The user’s tokens must be accessible to the settlement contract. This typically requires a one-time Permit2 approval, connecting approach 1 and approach 3.

How the Approaches Compare

In practice, these approaches are often combined. An intent-based system may use Permit2 for token approval. An account abstraction system may use Permit2 for spend authorization. The categories overlap.

Public Competitor Fees

When evaluating gasless swap options, it is worth understanding what competitors charge for their services. These are publicly documented fees:

• MetaMask Swaps charges a 0.875% service fee on every swap, factored into the quote rate. This fee has been unchanged since the desktop launch in October 2020. MetaMask uses a dynamic fee model that ranges from 0.3% to 0.875% based on order size.
• Matcha (powered by 0x) charges a 0.15% fee on its consumer interface. The underlying 0x API is free for integrators.
• 1inch charges 0% protocol fee on its aggregator. In Fusion mode (intent-based), infrastructure fees may apply, ranging from 0 to 5 basis points for stable pairs and 0 to 30 basis points for non-stable pairs, charged to the resolver rather than the user.
• Uniswap charges pool fees of 0.01%, 0.05%, 0.30%, or 1.00% depending on the fee tier. These are liquidity provider fees, not interface fees. Uniswap’s own interface does not charge an additional fee on top of pool fees.

These fees are separate from gas costs — for a full breakdown of all cost components in DEX trading, see our guide on understanding DEX aggregator fees. A gasless swap may eliminate the gas cost for the user but still carry an interface fee or a surplus capture by the solver.

Why This Matters for Base

Base is an Ethereum L2 built on the OP Stack. It uses ETH for gas. There is no native Base token. Every user who wants to transact on Base must hold ETH, even if the amount is tiny.

For users coming from centralized exchanges, this is a friction point. A user withdraws USDC to Base from Coinbase, wants to use it in DeFi, and cannot move it without ETH. Gasless swaps remove this barrier entirely.

The combination of Base’s low gas costs and gasless swap infrastructure makes it possible to onboard users who have never held ETH and never need to. They can bridge stablecoins, swap tokens, and participate in DeFi without ever acquiring the native gas token.

This is the thesis behind gasless swaps on Base. The technology exists today, through Permit2, account abstraction, and intent-based execution. The question is not whether gasless swaps are possible. They are. The question is which combination of approaches delivers the best user experience at the lowest cost.

Conclusion

Gasless swaps are not a single technology. They are a category of solutions that share a common goal: letting users trade without holding ETH for gas. Three approaches dominate the space. Permit2 reduces the number of on-chain transactions through off-chain signatures. Account abstraction uses paymasters to sponsor gas. Intent-based execution delegates the entire transaction to solvers who pay gas and compete on price.

Each approach has tradeoffs. Permit2 still requires the user to pay gas. Account abstraction adds infrastructure complexity. Intent-based execution introduces solver dependency. In practice, production systems combine elements from all three.

On Base, where gas costs are negligible, the economics of gasless swaps are particularly favorable. The cost of sponsoring gas or running a solver is small. The benefit to users is large. As the ecosystem matures, gasless swaps are likely to become the default rather than the exception.

If you want to see one implementation of gasless swaps on Base, Orkid is a product built on Base that offers gasless trading. How it works is beyond the scope of this article, which covers only the general concepts and public standards that make gasless swaps possible.

Frequently Asked Questions

What are gasless swaps?

Gasless swaps are token trades that do not require the user to hold the native gas token (ETH) to complete the transaction. Instead of paying gas directly, the user signs an off-chain message or relies on a third party to cover the gas cost. This removes the need to acquire ETH before trading.

Do you need ETH for gasless swaps on Base?

No. Gasless swap mechanisms allow you to trade without holding ETH in your wallet. A sponsor, paymaster, or solver pays the gas on your behalf, often recovering the cost through the swap itself or through a separate fee model. You only need the token you want to swap.

How do gasless swaps work on Base?

Base uses ETH for gas just like other EVM chains, but gasless approaches bypass this requirement through off-chain signatures, account abstraction paymasters, or intent-based execution where a third party submits the transaction. The user signs their intent off-chain, and another party handles the on-chain execution and gas payment.

Are gasless swaps safe?

Gasless swaps are generally safe when using reputable protocols, but they introduce trust in the party that submits your transaction. The cryptographic signature ensures your funds cannot be moved beyond your specified parameters. Users should verify they are interacting with legitimate interfaces and understand who is executing the transaction on their behalf.

Written by Jacob Cavazos