Gas Fees Across Chains: A Monthly Comparison

Published fee comparisons frequently mix measurement dates, use different operations, and convert to currency at inconsistent rates. This is the method we use, which anyone can reproduce.

The method

Three standard operations. A native token transfer, a token transfer, and a swap through the most liquid decentralised exchange on that chain.

Fixed sampling. Base fee or equivalent recorded at 08:00, 14:00 and 20:00 UTC, daily, for a calendar month.

Currency conversion at the moment of sampling. Not at the end of the month. A fee denominated in a volatile native token, converted at a single later rate, produces a chart of the token price rather than the fee.

Median rather than mean. A single congestion event distorts an average badly. The median describes what a user typically experiences.

What the figures consistently show

Ethereum mainnet is one to two orders of magnitude more expensive than everything else. This gap has been stable for years and has not narrowed.

Rollup fees are dominated by data costs, not execution. Since the introduction of dedicated data availability capacity, rollup fees fell sharply and have stayed low. The remaining variation between rollups is smaller than the variation within any one of them across a week.

Alternative layer 1 chains cluster together. Networks with higher throughput and lower decentralisation guarantees produce fees in a similar range to rollups.

Swap costs scale differently from transfers. A chain can be cheap for transfers and expensive for swaps if its execution pricing differs from its data pricing. Comparing only transfers misleads anyone who mostly swaps.

The columns worth building yourself

Chain	Native transfer	Token transfer	Swap	Sample period

Fill it for the chains you actually use, over a month, at fixed times. The absolute numbers age within weeks. The relationships between chains are stable for much longer, and they are what should drive a decision.

The trap in every published comparison

Most articles quote a single figure captured at one moment, frequently at a quiet hour, and present it as typical.

Mainnet fees vary by a factor of three to five across a week. A comparison using a Sunday morning figure for one chain and a Wednesday afternoon figure for another is not a comparison of chains.

Always check the timestamp. If there is not one, the figure is not usable.

What this does not measure

Security assumptions. A cheaper chain may have fewer validators, a smaller economic cost to attack, or an upgrade key held by a small group. Cost per transaction is one axis and not the important one for holding value.

Withdrawal cost and time. Getting off a chain is a separate cost, and on optimistic rollups a separate seven-day wait.

Liquidity. A cheap swap on a thin pool loses more to slippage than an expensive one on a deep pool. For anything beyond small amounts, depth dominates fee.

That last point is why we also track depth on the venues we reference. Fee comparisons that ignore slippage are comparing the visible cost and not the total one, and the venues that publish both, including an exchange with a published withdrawal schedule, make the full calculation possible rather than approximate.