Ethereum introduces per-transaction gas limit with fusaka upgrade to improve scalability

The Ethereum Foundation has officially confirmed that the upcoming Fusaka hard fork will introduce a strict upper limit on gas consumption per individual transaction, as defined in Ethereum Improvement Proposal 7825 (EIP-7825). This marks a significant shift in how Ethereum handles transaction execution, aiming to improve network stability and scalability by capping each transaction at 2²⁴ gas units—approximately 16.78 million gas.

This is the first time Ethereum enforces a per-transaction gas cap distinct from the overall block gas limit. Until now, while the total gas per block was limited (currently around 30 to 45 million gas), individual transactions could consume a disproportionate share of that capacity, creating potential issues for network performance and transaction predictability.

The new limit is already live on testnets Holesky and Sepolia and is expected to be implemented on the Ethereum mainnet with the activation of the Fusaka upgrade. According to core developer Toni Wahrstätter, who authored the informative post on October 21, the purpose of this change is twofold: to strengthen the network’s resilience against denial-of-service (DoS) attacks and to enable a more predictable and efficient transaction flow in preparation for future improvements like parallel execution.

By setting this cap, Ethereum aims to prevent oversized transactions from monopolizing entire blocks. In the past, complex smart contracts or large deployment scripts could consume nearly all available gas in a block, creating bottlenecks for other users and making block construction unpredictable for builders and validators. The new model forces developers to break down operations exceeding 16.78 million gas into smaller, sequential transactions, improving distribution and fairness.

Importantly, Ethereum Foundation reassures that this change will have minimal to no impact on the average user. Most day-to-day transactions fall well below the new cap. The groups most likely to be affected are developers deploying large contracts, executing batch operations, or using sophisticated routing mechanisms within DeFi protocols.

From a broader strategic standpoint, EIP-7825 lays the groundwork for Ethereum’s long-term scalability roadmap. It aligns closely with EIP-7928 and the anticipated “Glamsterdam” phase, which introduces parallel transaction execution. For parallelism to function efficiently, block composition must be predictable and transaction sizes manageable. The gas cap ensures that, even during extreme mempool congestion, multiple transactions can always be included in a block.

Technically speaking, the specification of EIP-7825 is simple: impose a hard ceiling of 16,777,216 gas per transaction. Developers and protocol researchers have praised this simplicity, noting that it enhances DoS resistance while remaining compatible with future scalability upgrades. The clarity of the rule has made it popular among core development teams.

Discussions leading up to this implementation were extensive. Core developers debated the best way to encode and enforce the limit across major Ethereum clients, including Geth, Erigon, Reth, Nethermind, and Besu—all of which have now integrated the change in their Fusaka-compatible versions. This unified client implementation minimizes the risk of chain splits or other inconsistencies at the time of the fork.

One of the key considerations was ensuring that block gas targets align with multiples of 2²⁴. This alignment guarantees that, as long as n transactions are available in the mempool that meet the criteria, at least n can be included in a block. This is less about maximizing throughput and more about ensuring consistency and reliability in block construction.

It’s worth noting that eth_call operations—used for querying the blockchain without altering state—will not be affected by the new gas limit. However, pre-signed transactions that exceed the cap will be rejected, and developers will need to adjust their dApps and deployment strategies accordingly.

Beyond the technical implications, EIP-7825 reflects a philosophical evolution in Ethereum’s governance. It demonstrates the network’s commitment to balancing decentralization with performance, and to building a platform that can support both complex applications and high transaction volumes without compromising security or predictability.

Looking ahead, this change could also influence gas pricing dynamics. By capping individual transaction gas usage, the network may see a more even distribution of fees across transactions, potentially reducing fee volatility during periods of high demand.

Another area where EIP-7825 could have an impact is on layer-2 solutions. These scaling technologies often aggregate multiple transactions into one, and the new gas limit may influence how rollups and zero-knowledge proofs are structured. Developers working on these solutions may need to optimize their compression techniques or adjust aggregation thresholds to ensure compatibility with the new transaction ceiling.

The introduction of a per-transaction gas cap could also make Ethereum more accessible to new developers. With clearer boundaries and simpler constraints, onboarding and tooling may become more straightforward, lowering the barrier to entry for building on Ethereum.

In addition, the change may foster innovation in transaction scheduling and optimization. As developers are encouraged to break down large operations into smaller components, new patterns and best practices are likely to emerge. This evolution in development style could lead to more modular and efficient smart contract architectures.

Finally, the Fusaka upgrade signals Ethereum’s readiness for a new chapter in its evolution—one focused not just on scalability, but on predictability, efficiency, and long-term developer experience. EIP-7825 is a relatively small change in terms of code, but its implications are far-reaching. It is a foundational step toward a more robust, flexible, and performance-oriented Ethereum.