Fusaka is a major network upgrade for the Ethereum blockchain that activated on mainnet on December 3, 2025. As the second major upgrade of 2025, following Pectra, Fusaka's primary objective was to enhance the network's scalability, efficiency, and security. The name is a portmanteau of Fulu and Osaka, representing the simultaneous upgrades to Ethereum's consensus and execution layers, respectively. The upgrade's developer-chosen mascot is a zebra, symbolizing the column-based data sampling method of its main feature, PeerDAS. This upgrade was designed to significantly reduce data costs for Layer-2 rollups and increase the network's overall throughput. ^{[17] [18] [28]}​

Overview

The Fusaka hard fork represented the next significant step in Ethereum's ongoing development roadmap, following the Pectra upgrade in May 2025. Its primary objective was to improve the overall performance and resilience of the Ethereum network by focusing on backend enhancements rather than direct user-facing features. A central component was the implementation of PeerDAS (Peer Data Availability Sampling), which was designed to make Ethereum's base layer more efficient at handling the vast amounts of data from its growing Layer-2 ecosystem. The upgrade bundled a set of Ethereum Improvement Proposals (EIPs) designed to optimize core protocol functions related to scalability, node health, and efficiency, leaving existing smart contracts untouched. ^{[19] [21]}​

Technology

The Fusaka upgrade is a comprehensive update that included 13 Ethereum Improvement Proposals (EIPs) aimed at enhancing the network's capabilities. The flagship feature was EIP-7594 (PeerDAS), which enhances data availability for L2s by allowing nodes to verify data by checking small, random samples ("slices") instead of downloading entire datasets. This enables a theoretical 8x increase in data space for blobs, significantly reducing costs for rollups. ^{[28] [19]}​

However, not all initially considered proposals were part of the Fusaka release. EIP-7907, which aimed to double the contract code size limit and introduce gas metering, was excluded to streamline the testing process and avoid potential delays. Similarly, the controversial EVM Object Format (EOF) upgrade, which proposed an overhaul of the Ethereum Virtual Machine (EVM) to simplify smart contract development, was officially removed from Fusaka due to technical uncertainties and community pushback regarding its complexity and implications. The decision to remove EOF was made to prioritize other critical features like PeerDAS and to ensure the upgrade's timeline was not jeopardized. ^{[14] [3] [4]}​

Another significant change included in Fusaka was the increase of the Ethereum gas limit. EIP-7935 was an informational EIP that coordinated client teams to raise the default block gas limit to approximately 60 million, up from the pre-Fusaka level of 45 million. This allows more transactions and computations to be processed in each block, enhancing network throughput and supporting more complex operations on the base layer. ^{[17] [28]}​

Key Ethereum Improvement Proposals (EIPs)

Fusaka bundled 13 infrastructure-level EIPs that refine scalability, improve efficiency, and harden the network without breaking existing contracts. The key proposals are categorized below.

Blob Scaling

• EIP-7594 (PeerDAS): The central feature of the upgrade, this EIP implemented Peer Data Availability Sampling. It allows nodes to verify data availability by downloading small, random "slices" of data blobs instead of the entire dataset. This lightens the load on most nodes and enabled a theoretical 8x increase in data throughput for rollups. ^[28]
• EIP-7892 (Blob Parameter-Only Forks): This created a lightweight framework for future adjustments to blob capacity (target and maximum per block) without requiring a full hard fork, streamlining subsequent scalability upgrades. ^[28]
• EIP-7918 (Blob Base Fee Bound): This EIP linked blob fees to execution costs by establishing a minimum reserve price for data blobs. This creates a fairer, more predictable pricing mechanism by preventing the blob base fee from collapsing to near-zero during periods of low demand. ^[28]

L1 Scaling & Security

• EIP-7642 (History Expiry): An informational EIP that mandated client support for history expiry. This built on the partial history expiry implemented in July 2025 to further reduce node storage and synchronization burdens, supporting long-term decentralization.
• EIP-7825 (Spam Resistance Checks): This proposal prevents malicious transaction spam by capping the gas limit for any single transaction at 16,777,216 (2^24). This helps nodes remain stable and enables safer future increases to the block gas limit. ^[28]
• EIP-7823 & EIP-7883 (MODEXP Adjustments): These two EIPs strengthened Ethereum’s resilience to denial-of-service (DoS) attacks. EIP-7823 capped the input sizes for the MODEXP precompile to a maximum of 1024 bytes, while EIP-7883 adjusted its gas pricing to better reflect its real-world computational cost. ^[28]
• EIP-7934 (RLP Execution Block Size Limit): This imposed an upper limit of 10 MiB on the size of RLP-encoded execution blocks to limit network bloat and reduce denial-of-service risks.
• EIP-7935 (Default Block Gas Limit): An informational EIP that coordinated client teams to raise the default block gas limit to approximately 60 million, enabling more transactions per block.

User Experience (UX) and Developer Improvements

• EIP-7951 (secp256r1 Precompile): This introduced a precompile for secp256r1 signature verification, enabling native support for passkeys and device-based signers (e.g., Apple Secure Enclave, WebAuthn). This paves the way for wallets to use biometrics like Face ID instead of seed phrases. ^[28]
• EIP-7917 (Deterministic Proposer Lookahead): This allows the Beacon Chain to precompute block proposers for the entire next epoch. This streamlines the validation process and enables L2 networks to offer features like near-instant transaction preconfirmations. ^[28]
• EIP-7939 (CLZ Opcode): This added a low-cost "count leading zeros" (CLZ) instruction to the EVM, which cheapens and optimizes certain bitwise and arithmetic calculations used in cryptography and compression. ^[28]

Meta EIPs

• EIP-7910 (eth_config RPC Method): An informational EIP for a new RPC endpoint allowing node operators to query fork configurations. This helps prevent misconfigurations that caused issues on testnets during the Pectra upgrade, ensuring smoother transitions. ^[28]

This set of EIPs was finalized after an EIP freeze on August 1, 2025, to ensure a focused and testable upgrade. [[cointelegraph.com/explained/ethereums-fusaka-upgrade-set-for-november-what-you-need-to-know][Cointelegraph explanation of Fusaka upgrade]] [[etherworld.co/2025/06/17/all-you-need-to-know-about-ethereum-fusaka-upgrade/][EtherWorld overview of Fusaka]]

Development and Timeline

Preparations for the Fusaka hard fork were actively underway throughout 2025, following an aggressive development schedule. The process involved client teams preparing their software for the upgrade and multiple development networks (devnets) to test and refine the included EIPs. After an initial devnet, Fusaka Devnet 2 launched on June 27, 2025, serving as a "bug-hunt" to identify issues, followed by Devnet 3 to test the "happy-path" scenario. A final devnet was launched on July 23, 2025, with public testnets following in October. ^{[1] [22] [15]}​

Despite the ambitious timeline, some members of the Ethereum community had expressed concerns about meeting the proposed launch window. Nixo, an Ethereum protocol support member, noted the tight schedule required to ship the upgrade before Devconnect. However, developers successfully conducted all planned tests, leading to a finalized mainnet date. ^{[7] [5]}​

Testnet and Mainnet Launch

The testnet deployment for the Fusaka upgrade was successfully completed in October 2025 across three "dress rehearsals." The upgrade was activated on Holesky (October 1), Sepolia (October 14), and finally on Hoodi (October 28), concluding the testnet phase and confirming the upgrade's readiness for mainnet. ^{[23] [24]}​

On November 12, 2025, the Ethereum Foundation formally announced the mainnet launch date. The Fusaka hard fork activated on December 3, 2025, at 21:49:11 UTC, corresponding to block 13,164,544. The launch date was finalized by core developers on October 30, following the successful testnet deployments, placing it after the Devconnect conference held in Buenos Aires in mid-November. ^{[25] [26] [6]}​

Following the main fork, two smaller "Blob Parameter Only" (BPO) forks are planned to increase blob capacity:

• BPO Fork 1: Scheduled for December 17, 2025, this fork will increase the maximum blob count per block from 9 to 15.
• BPO Fork 2: Scheduled for January 7, 2026, this fork will further increase the maximum blobs per block to 21.

These lightweight forks are designed to scale network data capacity without requiring full client updates. ^{[17] [20]}​

Security Audit Contest

To bolster security ahead of the mainnet launch, the Ethereum Foundation announced a 100,000 and Lido contributed $25,000 to the prize pool. The competition invited security researchers to identify vulnerabilities in the upgrade's codebase, with a 2x points multiplier for valid findings in the first week and a 1.5x multiplier in the second week to incentivize early discoveries. ^{[16] [20]}​

Impact on Stakeholders

Impact on Users and Developers

For developers, Fusaka was designed to be a non-disruptive, backwards-compatible upgrade. Existing smart contracts were not expected to break, though contracts with very high gas consumption (approaching the new 16.7 million transaction gas cap) or heavy MODEXP precompile usage may have required optimization. The primary benefits for builders come from the increased gas limit, which allows for more complex operations, and performance improvements from PeerDAS, which lowers data fees for rollups. New tools like the CLZ opcode and native passkey support (secp256r1) provide new tools for optimizing on-chain logic and creating user-friendly wallets. ^[28]​

For everyday users, the main benefit of Fusaka is expected to be lower transaction fees on Layer-2 networks. By making it cheaper for rollups to post data to Ethereum, the upgrade should lead to more affordable user transactions on L2s. The upgrade was not expected to significantly lower gas fees on the Ethereum mainnet itself. Additionally, the secp256r1 precompile paves the way for a new generation of wallets secured by device biometrics (e.g., Face ID) instead of seed phrases, simplifying user onboarding and account recovery. ^{[18] [28]}​

Impact on Node Operators

All node operators and stakers had to update their execution and consensus client software to Fusaka-compatible versions to remain synced with the network. The hardware impact varies by operator type: ^[28]​

• Full Nodes (Non-validating): Experienced a significant decrease (around 80%) in bandwidth and storage needs related to blobs, as they now only need to store a fraction (approx. 1/8th) of the total blob data.
• Solo Stakers: Saw a moderate decrease in download bandwidth. However, those who build blocks locally may require higher upload bandwidth as blob capacity increases via subsequent BPO forks.
• Large Validators / Supernodes (>= 4096 ETH): These operators, who now custody all blob data, saw a substantial increase in storage and bandwidth requirements as blob capacity scales. They play a key role in "healing" the network by providing full data copies.

Future Upgrades

Following Fusaka, attention is already turning to the subsequent Ethereum upgrade, codenamed Glamsterdam, which is anticipated in 2026. While plans are not finalized, developers intend to include enshrined Proposer-Builder Separation (ePBS) in the upgrade. Another proposal, put forth by Ethereum core developer Barnabé Monnot, suggests halving Ethereum's block time from 12 seconds to six seconds to enhance the performance and user experience of decentralized applications (dApps). ^{[23] [8] [2]}​

Regulatory Discussions

In parallel with the technical developments, the U.S. Securities and Exchange Commission (SEC) engaged in discussions in 2025 with key blockchain industry stakeholders, including representatives from Ethereum-aligned organizations such as the ERC-3643 Association, Chainlink Labs, the Enterprise Ethereum Alliance, and LF Decentralized Trust. The meeting focused on exploring potential token standards that could facilitate the compliant issuance and transfer of tokenized securities.

The discussions primarily revolved around integrating open standards like ERC-3643 and compliance frameworks such as Chainlink’s Automated Compliance Engine (ACE) into the existing regulatory landscape. ERC-3643 is a token standard designed for compliant capital markets on Ethereum, while Chainlink’s ACE provides a smart-contract-based compliance framework for managing tokenized assets, including securities and real-world assets.

Dennis O’Connell, president of the ERC-3643 Association, indicated that the meeting signaled a notable shift in the SEC's approach, demonstrating increased engagement and a willingness to understand how blockchain standards can align with regulatory oversight. He noted that the discussions helped illustrate how standardized frameworks could enable regulatory compliance while fostering innovation.

Following the meeting, SEC Chair Paul Atkins suggested that the agency is considering an "innovation exemption" within its regulatory framework to support tokenization. Atkins acknowledged the inevitable migration of assets onto blockchain networks and indicated that the SEC might implement changes to accommodate new trading methods tailored for tokenized securities. ^{[9] [10] [11] [12]}​

Economic Considerations

The Fusaka upgrade has several economic implications. The increase in the Ethereum block gas limit to 60 million, while intended to improve throughput, sparked concerns that higher hardware demands could strain smaller solo validators and nudge the network toward centralization. Similarly, while PeerDAS lowers the barrier for most nodes, its design may lead to the emergence of "supernodes"—highly capable nodes run by large entities that store the full L2 data. Critics worried this could centralize the critical function of data storage, though proponents argued this was a safe trade-off since all validators can still independently verify data availability via sampling. ^{[28] [17]}​

Anticipation for the upgrade was a key factor in market activity. In the months leading up to November 2025, large investors ("whales") accumulated over 7.6 million ETH. However, analysts noted a potential trade-off: by creating dedicated "blob lanes" for L2 data, Fusaka is expected to lower fee pressure on the base layer. This could reduce the amount of ETH burned via EIP-1559, potentially weakening Ethereum's deflationary monetary policy. ^{[27] [26]}​

Analysts viewed the upgrade as a critical development for maintaining Ethereum's competitive position as the primary settlement layer for decentralized applications, especially as capital flows to rival blockchains. A successful implementation was expected to reinforce Ethereum’s market dominance by enhancing scalability and reducing transaction costs for the entire L2 ecosystem. ^[13]​