AIP: Raise the gas target, min L2 base fee, & implement improvements to the pricing algorithm

[gi0rgos]

Thank you for the proposal!
This is an important topic, as changes in gas fees can affect different types of users in different ways, especially with this climax structure.

I understand the change proposed by @Entropy and also the concerns raised by @Jonezee regarding the potential impact on everyday and smaller users. I have spent some time trying to analyze the real effects, since losing users from the ecosystem would be a serious issue. Based on what I have found so far, while there may be some impact for certain use cases, I believe that the overall benefits outweigh the drawbacks. Therefore, at this stage, I intend to vote FOR.

Of course, if anyone has concrete data showing that a minimum of 0.02 gwei would harm users and, in conclusion, the ecosystem, please share it with the community so we can evaluate it before the on-chain vote.

[shogun]

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 are generally in accordance with the increased minimum L2 base fee, its important to note that the aforementioned spike in October 10th 2025 and September 22nd 2025 in gas expense, and more recently 4th November 2025, was related to a series of perps orders being triggered. Perps and derivatives users on Arbitrum, will unambiguously feel this expense, what we hope is that this doesn’t deliver an undesirable result for traders to move away in high demand usage.

Nonetheless, we’re excited for the technical upgrades for node operators and the Erigon team’s deliverable to the Sepolia network and soon to Arbitrum One for these operators in reducing storage and costs.

We have APPROVED this proposal, but hope there’s a recognition for trader behaviour.

[offchainlabs]

Hey dzack23, great question. The reason is so that future changes may involve the introduction of a gas target for a very short window of 5 seconds (almost half of the current shortest adjustment window of 9s). A gas target measured over a 5 second adjustment window would be used to allow the chain to handle extreme demand shocks for a brief, short period of time. For example, a gas target of 75Mgas/s over 5 seconds may be accompanied by gas targets measured over longer periods of time (than the ones already proposed).

[maxlomu]

@Jonezee

In quiet times, users pay about 7 cents, but that can easily go up to 20 cents.

Hey @Jonezee thanks for bringing up your/ GMX’s perspective. I believe the purpose of this proposal is exactly to reduce those spikes in times of high congestions.

[Camelot]

Camelot supports this proposal as we see value in having a smoother fee curve during periods of network congestion. We have observed what happens during turmoil periods, and we believe the proposed changes would benefit user experience by reducing the volatility in gas pricing, especially in DeFi protocols that do see increase of usage like DEXes and Perps.
However, we are somewhat dubious about increasing the base fee. While there is certainly merit in both increasing base revenue for the DAO and eliminating spam transactions, we wonder how this would affect arbitrageurs and market makers. We are particularly interested in understanding if there are any projections on how this increase might change activity that is key for specific DeFi protocols, namely DEXs and perpetual protocols.

We are overall supportive of the direction, but before moving to Tally we would like the question above answered. We acknowledge there is a scenario in which market maker data is simply not public, not even for Offchain Labs, and so conducting projections might be complex. If such data constraints exist, we would appreciate transparency around that limitation and any alternative analysis that could be provided to assess the potential impact on arbitrage and market-making activities.

[offchainlabs]

Thanks for the thoughtful questions and perspectives @Camelot @gi0rgos @Jonezee @shogun

It’s difficult to produce precise projections for how a 2x increase to the minimum base fee may influence arbitrage, market-making, and user flows, holistically - these behaviors adapt to many variables. However we can lean on historical usage - the minimum base fee was previously 10x higher (0.1 gwei) before the Arbitrum Security Council was requested to take a non-emergency action to reduce it to 0.01 gwei in March 2024; activity across DeFi protocols (including DEXs and perps) remained healthy before that reduction.

The intent behind the proposed fee curve changes is to smooth out the user experience of fees during high-activity events, while the intent behind the related proposed 0.02 gwei minimum is to not meaningfully impact typical user experience but to balance and stabilize DAO revenue as the smoother fee curve reduces surplus fees. At the same time, the adjustment in the base fee helps make spam-style bot activity more expensive, which causes unpredictable fees for organic users. The adjustment remains well within historically permissible parameters for the ecosystem.

For a deeper explanation of the rationale, we encourage reading this response from @Entropy

Entropy Advisors is strongly in favor of this proposal. As always, thank you to the Offchain Labs team for putting this together.

However, we are in favor of modifying this proposal to also include a 2x of the minimum base fee from .01 gwei to .02 gwei. We wrote extensively about this topic and our rationale behind our view point here.

To quickly summarize, we estimate that roughly 80-90% of daily transactions conducted on Arbitrum One are bot-driven in nature. While increasing the minimum base fee has a minimal effect on “average” users that engage with applications in an organic manner, it will increase the costs of transacting for bot operators. We view this as positive, as it could help prevent spam-motivated bots that congest the network for ordinary users, and could even push arbitrage bots to engage with Timeboost and make the product a more competitive marketplace.

Screenshot 2025-11-05 at 12.36.31 PM721×406 49.5 KB

As a side benefit, a slight increase to the L2 min base fee will also ensure the DAO does not experience a decrease in revenue, while also improving the revenue predictability and thus financial forecasting/budgeting capabilities of the DAO. As can be seen in the image below, L2 Surplus Fees generate a lot of revenue for the DAO but are more sporadic in nature. This leads to a less than ideal UX for organic users on Arbitrum One during times of congestion, which is why we are in support of this proposal from OCL. It is simply not worth the added revenue that accrues to the DAO.

Screenshot 2025-11-05 at 12.49.56 PM721×406 21.7 KB

In contrast to L2 surplus fees, the L2 base fee provides a much more stable source of DAO revenue (image below) and does not come at the expense of organic users’ experience. The L2 min base fee on Arbitrum One was previously set to 0.1 gwei prior to the ArbOS Atlas upgrade, so we already know that this is a safe change to make.

Screenshot 2025-11-05 at 12.52.22 PM721×404 23.5 KB

Over the past 90 days, the DAO has generated ~$6.7M of revenue from transaction fees, of which only $1.8M was derived from the L2 base fees and the remainder coming from L2 surplus fees. With this proposal, the L2 surplus fees are set to be drastically reduced. However, if we increase L2 min base fee from .01 gwei to .02 gwei, paired with the increased gas target per second and subsequent increase in daily transaction count from bots, the min base fee increase should help offset the revenue decline in L2 surplus fees.

To sum it up - we are in favor of this proposal as is, but believe it is made stronger by incorporating a slight increase to the L2 min base fee. This has the potential to increase Timeboost participation, make spam activity more expensive to conduct (thus providing more, cheaper block space for organic users), and offset the decline in L2 surplus fee-derived revenue by increasing L2 min base fee-derived revenue which is more predictable in nature.

[gi0rgos]

Thank you for the comment.

I had studied the Entropy position in detail before posting my rationale, and I agree with you that it provides a solid rationale for the adjustment.

However, what I did not know, and is another strong argument, is that in the past, the minimum fee level was 10 times higher, and yet no harm to protocol activity was observed. This is even more reassuring.

Thank you for offering this perspective on information I was not aware of. It helps me form an even clearer view on the matter.

[paulofonseca]

voting For on this offchain vote because this will improve the network

[Michigan_Blockchain]

Your minimum fee level point was a good one, I didn’t even think of that. Definitely in favor of the proposal.

[cp0x]

Voted FOR

Cp0x Delegate Communication Thread

AIP: Raise the gas target, min L2 base fee, & implement improvements to the pricing algorithm The minimum fee for L2 calculations is being increased from 0.01 to 0.02 gwei. At the same time, the threshold for automatic gas price increases is being increased (it increases during peak loads). This will not significantly increase transaction costs during normal times, but will prevent sharp price spikes during peak loads - an excellent solution for both Arbitrage and users. Offchain Labs also h…

[Zeptimus]

Voting For on Tally

Having more predictable fees makes sense in all scenarios. Even as a regular user, knowing roughly what you’ll pay is valuable. But it’s especially important for the protocol because predictable pricing is what allows developers to build sustainable business models on top of Arbitrum.

Raising the minimum L2 base fee to is a small and low risk change. Fees on L2s are already very low, so general users won’t even notice .

As for node operators, I don’t think this change will actually reduce the number of people running nodes. Operators who have economic or strategic reasons to run nodes will adapt. The impact on network participation should be limited.