[CONSTITUTIONAL] AIP: Update Arbitrum One and Nova L1-Pricing Parameters for Ethereum Glamsterdam

[CONSTITUTIONAL] AIP: Update Arbitrum One and Nova L1-Pricing Parameters for Ethereum Glamsterdam

Abstract

This proposal updates the parentGasFloorPerToken and perBatchGasCharge parameters on Arbitrum One and Arbitrum Nova in response to Ethereum’s Glamsterdam upgrade.

Glamsterdam changes the Ethereum gas costs associated with posting Arbitrum chain batches. Without corresponding parameter updates, the L1 fee paid by Arbitrum One and Nova users may no longer accurately reflect the cost of posting batches to Ethereum.

The proposal authorizes targeted updates to the live parameters through the ArbOwner precompile. It does not introduce an ArbOS software upgrade or change the underlying L1-pricing algorithm.

Motivation

Arbitrum’s L1-pricing system estimates Ethereum gas cost of each batch, then uses that estimate to set the L1 fee paid by Arbitrum users and the amount the batch poster is paid back. Two parameters feed into this estimate:

  • parentGasFloorPerToken, which mirrors Ethereum’s minimum gas charge for calldata; and
  • perBatchGasCharge, which represents a fixed component of the estimated Ethereum gas cost attributed to each batch.

The existing values reflect Ethereum’s pre-Glamsterdam cost structure. Following Glamsterdam’s activation, they may no longer accurately account for the costs incurred by the Arbitrum One and Nova batch posters.

This proposal updates the parameters to reflect the expected post-Glamsterdam posting costs.

Rationale

Updating these two parameters is the smallest change that keeps the L1 fee paid by Arbitrum users in line with what the batch posters spend after Glamsterdam. It uses the existing ArbOwner controls and requires no change to Nitro software, ArbOS or the L1-pricing algorithm.

The proposed values are based on measured post-Glamsterdam batch costs. They may be revisited if posting costs or the L1-pricing system changes.

Specifications

Affected chains

This proposal applies to:

  • Arbitrum One; and
  • Arbitrum Nova.

It does not change parameters on Arbitrum Sepolia or independently governed Arbitrum chains.

Parameter updates

Parameter Current value Proposed value
parentGasFloorPerToken 10 16
perBatchGasCharge 210,000 480,000

Note: these values may underestimate the gas paid for zero-bytes posted for calldata. The effect is minimal, since zero bytes are rare and Arbitrum One and Nova post blobs. A future ArbOS version could address this.

For each chain, the Governance Action Contract will call:

ArbOwner.setParentGasFloorPerToken(16);
ArbOwner.setPerBatchGasCharge(480000);

The proposal also includes an onchain check (GlamsterdamForkGateAction) that blocks execution until Glamsterdam is live on Ethereum. Any earlier attempt reverts, and the proposal can only be executed once the fork is active.

Implementation work is available in ArbitrumFoundation/governance PR #384.

Expected Impact

If approved and executed, the proposal is expected to improve the alignment between:

  • The Ethereum gas costs incurred by the Arbitrum One and Nova batch posters after Glamsterdam; and
  • The reimbursement calculated by the existing ArbOS L1-pricing system.

Updating these parameters will affect the L1 fee component paid by users.

The proposal does not change Nitro node software, transaction-execution rules or the ArbOS version running on either chain.

Steps to Implement

  1. Forum discussion: An AIP outlining the proposed parameter changes is posted to the ArbitrumDAO Forum for discussion (this post).
  2. Temperature check: Submit the proposal to Snapshot.
  3. Implementation and review: The governance payload is finalized, tested and audited by a third party. The audit report, source code and deployment addresses are published before the onchain vote.
  4. Onchain vote: Following a successful temperature check and completion of the implementation materials, submit the Constitutional AIP for an on-chain vote.
  5. Governance execution: If approved, complete the required governance waiting periods and execute the proposal once Glamsterdam is active on Ethereum.

Overall Cost

This proposal comes at no additional cost to the ArbitrumDAO; the engineering work and coordination for an audit will be managed by Offchain.

1 Like

Entropy has posted this proposal to a temperature check vote on Snapshot on behalf of Offchain.

We support updating the L1-pricing parameters on Arbitrum One and Nova for Glamsterdam and look forward to the audit and final payload before the onchain vote.

1 Like

I support updating Arbitrum One and Nova’s L1-pricing parameters to reflect Ethereum’s Glamsterdam cost changes. Accurate pricing helps align user fees with actual batch-posting costs and appropriate reimbursement. The limited scope and execution check preventing activation before Glamsterdam are sensible safeguards.
Before final onchain approval, would appreciate published audit and testing to substantiate the proposed values and implementation.

We support the temp check. We decoded the payload data in PR #384 and it matches the proposal text.

The new perBatchGasCharge rests on a batch cost of ~353k gas after Glamsterdam, which the comment in SetGlamsterdamGasParamsAction says assumes no gas refunder, while today’s ~171k on Arb One is measured with one. Before the onchain vote, we would like confirmation that Arbitrum One’s batch poster will run without the gas refunder by the time this executes, and roughly how much of the 2.07x increase comes from that versus EIP-8037. It would help to have this dependency and the measurement data described in the onchain proposal itself.

Voting FOR. Makes sense to keep fees in line with what posting to Ethereum actually costs after Glamsterdam, and it’s a small change. I trust Offchain Labs execution on these.

Building on cp0x’s comparability concern, I think a four-cell reconciliation would help delegates evaluate the proposed L1-pricing adjustment.

The current Arbitrum One batch-cost measurement is approximately 171,000 gas with a gas refunder, while the proposed post-Glamsterdam estimate is approximately 353,000 gas without one. Comparing these figures directly combines a change in refunder assumptions with the modeled impact of Glamsterdam.

Could Offchain Labs publish a 2×2 comparison showing:

  1. Current batch cost, with refunder

  2. Current batch cost, without refunder

  3. Proposed post-Glamsterdam batch cost, with refunder

  4. Proposed post-Glamsterdam batch cost, without refunder

Ideally, the calculations would use consistent batch samples, identify the relevant EIP-8037 assumptions, and clearly distinguish observed measurements from modeled estimates.

This would allow delegates to compare the changes under consistent refunder conventions, identify the refunder effect separately, and assess any interaction between the two.

Where applicable, the same reconciliation for Nova would be useful.

The objective is not to challenge the need for updated pricing, but to ensure that the proposed parameter change can be evaluated against comparable measurements before the onchain vote.

Merlyn labs is voting FOR this proposal as well. It’s a small update that keeps fees matched to actual Eth costs after the Glamsterdam update.

Tied to successful testing and a completed audit before the onchain vote, we support the parameter changes in order to keep Arbitrum One and Nova L1 fees and batch-poster reimbursement aligned after Glamsterdam.

The governance payload is finalized, tested and has been audited. The report can be found here - https://docs.arbitrum.io/audit-reports/2026_10_19_trail_of_bits_glamsterdam_gas_parameters_summary_report.pdf

Thanks for the update @offchain . A narrowly scoped adjustment to Arbitrum One and Nova’s gas parameters makes sense if it keeps L1 cost reflection and batch-poster reimbursement aligned after Glamsterdam. The testing and audit update is helpful context; publishing the results and monitoring the parameters after activation would also help delegates assess the change over time.

Voting FOR… easy yes. Users should pay what posting to Ethereum actually costs us, I still think txs on Arbitrum are too cheap anyway. Props to Offchain for staying on top of this, as a delegate it’s hard to do a real deep dive into these technical upgrades.

Honestly for gas pricing updates like this one I would lean towards making some council or something and have optimistic approval…

I think it would be good to have someone actively adjusting the levers for our main revenue stream…

Also it would save the DAO a lot of time and resources. These votes cost real money and attention, no one is going to overrule Offchain unless they really overstep.

gm, voted FOR. Logical change, thanks OCL.

Cornell Blockchain is voting FOR this proposal as it is a logical and necessary update to keep Arbitrum One and Nova’s L1 fee calculation aligned with Ethereum’s Glamsterdam upgrade.

The following reflects the views of L2BEAT’s governance team, composed of @krst and @Manugotsuka, and is based on their combined research, fact-checking, and discussion.

We voted FOR.

With Glamsterdam changing the cost of posting batches to Ethereum, it makes sense to update Arbitrum One and Nova’s L1 pricing parameters accordingly. The goal here is fairly straightforward: keep the fee model aligned with the actual costs the chains are expected to face.

We also want to echo a point raised by other delegates around the new per-batch gas charge, as we think it is a fair concern. The proposed value assumes that Arbitrum One will no longer be using its gas refunder by the time this change is executed. Nova is already operating without one, but for Arbitrum One this still depends on a separate configuration change that is not part of the proposal itself.

At this stage, we did not find any blockers that would prevent us from supporting the proposal. We would still prefer to do the deeper technical review once it reaches the onchain stage and the final executable is available. At that point, we plan to ask our Research Team to go through the payload and verify that the implementation still matches the assumptions discussed during the offchain process.

1 Like

The following reflects the views of GMX’s Governance Committee, and is based on the combined research, evaluation, consensus, and ideation of various committee members.

We voted FOR on Snapshot. Glamsterdam changes the cost of posting batches to Ethereum, so the L1 fee users pay must move with it. The change itself is small and needs no ArbOS or Nitro upgrade. A fork gate reverts the whole batch until Glamsterdam is live, and Trail of Bits reviewed the gate, values, and payload and found no security issues.

1 Like

MUX voted FOR this proposal. he update is relatively minor and doesn’t require changes to ArbOS or Nitro. and Trail of Bits has reviewed the implementation without identifying any security concerns.

Reverie is voting FOR this proposal. In the unlikely scenario where both Arbitrum’s L2 governance contracts and Security Council are unable to act, we think it’s important that tokenholders retain a way to recover governance without relying on either.

I support this proposal, consistent with my temperature-check position. The published independent audit and testing provide reassurance on the implementation and the safeguard against execution before Glamsterdam.

It would also be helpful to confirm that Arbitrum One’s gas-refunder configuration will match the pricing assumptions before execution.

Thanks @cp0x and @martineza79.

Yes, the gas refunder is expected to be retired from Arbitrum One once Glamsterdam reaches mainnet. Removing it is a batch poster config change, so it isn’t in the payload. The fork gate stops the update from executing before Glamsterdam.

With refunder Without refunder
Current 171,016 144,034
Post-Glamsterdam 385,756 353,374

The refunder isn’t behind the increase. It adds about 27-32k per batch, so removing it lowers cost. The 171k to 353k jump is about +209k from Glamsterdam (~196k of that is EIP-8037 state-creation gas) minus ~27k from dropping the refunder.

Voting for onchain, same stance as before