ArbOS Dia: Smoother Fees, Higher Throughput

Most protocol upgrades are easy to describe to engineers and hard to feel as a product.

Arbitrum’s Dia upgrade is the opposite.

It’s the kind of upgrade that quietly changes the default experience: how users engage with daily gas fees, how operators think about stability, how developers make onboarding familiar, and how Arbitrum evolves alongside Ethereum without accumulating protocol debt.

This article focuses on what matters most: benefits and outcomes.

The big idea

The Dia upgrade brings Arbitrum into the next generation of Ethereum-aligned execution changes while advancing four product narratives that matter to companies shipping at scale:

• Smoother gas prices on Arbitrum One & Nova
• A foundation for more throughput with similar hardware
• Mobile & enterprise-grade authentication tools
• Continued alignment with Ethereum’s Fusaka-era EVM

If you build apps, run nodes, or care about user growth, this upgrade lands right in your world.

Why Dia matters now

As Arbitrum continues to grow, what matters most is that growth feels smooth and predictable for everyone using the network. Most builders want three things from the chain they deploy on:

• Onboarding that feels familiar
• Scale that feels seamless for users
• Gas behavior that is more predictable

Dia is designed to deliver all three.

Benefit for users: smoother gas prices

Anyone who has used a blockchain has felt this pain: the network gets busy, fees jump, transactions get delayed or dropped, and the experience gets less predictable.

Dia includes pricing updates for Arbitrum One to reduce the severity, frequency, and duration of high L2 gas prices during elevated demand.

Concretely, this includes improvements to the pricing model:

• Replacing a single gas target and adjustment window with a model that uses multiple (higher) gas targets across multiple (longer) adjustment windows, and
• Increasing the default minimum L2 base fee from 0.01 gwei to 0.02 gwei. This helps make spam-style bot activity more expensive, as well as balance and stabilize DAO revenue for the changes from a smoother fee curve.

The goal here is not to change what Arbitrum is. It’s to make high-demand moments feel less like rideshare surge pricing.

Benefit for operators: Similar hardware, more throughput

When a network grows, operators experience the pressure first.

Dia advances two operator-critical improvements that help the network behave more predictably under load and lay the groundwork for higher, more sustainable throughput.

The foundation for dynamic pricing

Dia implements Arbitrum’s State Transition Function to track gas usage across multiple resource types, including computation, storage access, storage growth, and history growth,  instead of just a single combined total. In other words, the chain starts to understand what is actually expensive, not just how much gas was used.

That sounds subtle, but it’s expected to be a big unlock for the future. This is the foundation for dynamic, constraint-based pricing, where fees can respond to the most constrained resource at the network level. Done right, that’s how you safely push much higher throughput (like 10x higher gas targets), keep node operators within their comfort zone, and make fee behavior more stable instead of spiky when usage patterns shift.

The aforementioned changes to smooth out gas price spikes allow the chain to absorb and handle more demand than before. Though these constraints are not enabled just yet, Dia’s role is to wire in the sensors: measure per-resource usage, validate the data, and only then turn the dials in a later ArbOS upgrade once constraints are configured, benchmarked, and tested. 

More efficient block packing

Because a per-transaction limit is introduced, the state transition rules are relaxed to allow the final transaction in a block to use up to the MaxTxGasLimit even if it nudges beyond the prior MaxBlockGasLimit.

This means the sequencer can pack blocks more efficiently without changing overall gas targets or the chain’s throughput policy.

Translation: fewer “why was my transaction skipped?” surprises, and a more stable experience during spikes.

Benefit for builders: mobile & enterprise onboarding

The blockchain industry has spent years teaching users a new mental model: seed phrases, backup rituals, and new security assumptions.

Though support for phone-native onboarding with secp256r1 was introduced in 2024, Dia upgrades and aligns that support with the version planned for Ethereum post-Fusaka. 

Practically, this means teams can treat passkey-style and biometrics-first signing as a more stable, long-term primitive on Arbitrum, opening the design space for apps and wallets to offer an experience users already understand: 

• Passkey-style authentication
• Face ID / fingerprint-style signing patterns
• Device-secured keys (including mobile keystore-backed flows)
• Corporate IDPs, 2FA, and account backup & recovery flows built on top

Arbitrum chains: more gas token types

Dia further supercharges Arbitrum chains by adding more flexibility around native gas tokens, strengthening the “Your Chain, Your Rules” model for builders. 

With Native Token Mint/Burn, Arbitrum chains gain support for interoperability-enabled gas token standards, including LayerZero OFTs, xERC20s, native USDC, and native USDT (USDT0), by allowing a chain to delegate minting and burning of its native gas token to a trusted bridge provider rather than relying only on the canonical “lock and mint” model.

The payoff is practical: cleaner compatibility with third-party cross-chain adapters, a wider gas-token design space, and reduced need for future forks by streamlining development into a unified codebase.

There are no plans to enable this feature on Arbitrum One. It’s a targeted unlock for custom Arbitrum chains, exactly where it belongs.

Benefit for the ecosystem: Ethereum alignment with Fusaka

Dia reinforces Arbitrum’s role as Ethereum-aligned infrastructure, with a familiar, predictable EVM experience for developers. 

Dia directly connects to Arbitrum's core values to build a:

• High-performance execution environment
• Forward-compatible ecosystem for apps that expect Ethereum equivalence
• Safer place to build as upstream cryptographic and gas semantics changes

In practice, Dia pulls in the Fusaka-era execution changes that matter for Arbitrum chains. That includes:

• Updating secp256r1 behavior (EIP-7951) so passkey-style flows match Ethereum’s spec
• Adding the new CLZ opcode (EIP-7939)
• Tightening and repricing the ModExp precompile (EIP-7823 and EIP-7883)
• Properly enabling BLS12-381 curve operations (EIP-2537)

On the node side, Dia also adds support for the eth_config RPC method (EIP-7910) and incorporates the relevant Fusaka networking and history cleanups where they apply.

At the same time, Dia deliberately leaves out Fusaka EIPs that target features Arbitrum doesn’t use today (such as blob data markets, beacon-chain P2P behavior, or L1 block propagation), and keeps options open around things like contract size limits as Ethereum’s own roadmap evolves.

The bottom line 

Dia is not an upgrade for people who collect EIPs. It’s an upgrade for people who ship products. It brings Arbitrum closer to the world where:

• Gas behavior is smoother when demand surges
• Operators get superior foundations for scaling without surprise hardware stress
• Onboarding can feel like using a modern phone or enterprise authentication, not passing an initiation ritual
• Ethereum alignment remains a feature, not a burden
• Custom chains get more design freedom with less fork fatigue

If Arbitrum’s long-term mission is to scale Ethereum without sacrificing the experience, then Dia is exactly what progress should look like: quietly ambitious, practically useful, and designed for the next wave of real users.

Ready to uplevel your blockchain infrastructure? Reach out at arbitrum.io/contact.