The next major upgrade for the Ethereum network is on the horizon. Called Fusaka—short for “Fulu-Osaka”—the release is scheduled for Q4 2025 and will combine significant changes in both Ethereum’s execution and consensus layers.
Fusaka follows several milestones for the Ethereum network following The Merge in 2022. Shanghai/Shapella in 2023 introduced staked ETH withdrawals, Dencun in 2024 added proto-danksharding and blobs, and Pectra in 2025 brought validator flexibility and layer-2 interoperability.
According to the project’s roadmap, Fusaka is designed to expand data capacity, reinforce defenses against denial-of-service attacks, and introduce new tools for developers and users.
The changes are sweeping. Fusaka is not a minor patch but a redesign of how Ethereum manages data availability, blob pricing, and transaction safeguards. Its success will be measured by whether the network can scale to meet rising layer-2 demand without fracturing or overburdening node operators.
PeerDAS: Sampling instead of storing everything
The central feature of Fusaka is PeerDAS, short for “data availability sampling,” a new way of handling blob data.
In Ethereum, a blob is a temporary data package introduced with proto-danksharding as part of the Dencun upgrade. Blobs let layer-2 rollups post large amounts of transaction data to the mainnet cheaply, improving scalability without permanently bloating the blockchain’s state.
That ensures redundancy—but creates a bottleneck as demand grows. In the current model, every full node on Ethereum must store every “blob” of layer-2 data posted to the chain.
PeerDAS changes the equation. Each node will store only a fraction of blob data—about one-eighth—and rely on cryptographic reconstruction to fill in missing pieces. The design relies on random sampling to verify data availability with extremely low error probabilities, on the order of one in 10²⁰ to 10²⁴.
By distributing storage this way, Ethereum can, in theory, support up to eight times more blob throughput without demanding higher hardware or bandwidth from node operators. Rollups, which depend on blobs to publish compressed transaction data, are expected to benefit most directly.
Blob economics and flexibility
Fusaka also reshapes how blob data is priced and managed.
One key change, EIP-7918, introduces a reserve fee for blobs. Under current rules, blob prices can collapse close to zero when execution gas fees dominate. This creates incentives for inefficient use. The reserve fee ensures that blob usage always carries a baseline cost, making Layer 2 networks pay for the storage and bandwidth they consume.
Another mechanism, EIP-7892, introduces blob parameter-only forks. These allow Ethereum clients to adjust blob throughput outside of full hard forks. The goal is to give developers the agility to respond to unpredictable Layer 2 demand without waiting for the next scheduled upgrade.
Guarding against attacks
Scaling up also means increasing Ethereum’s attack surface. Fusaka includes a series of changes to limit worst-case scenarios and protect the network from denial-of-service attacks:
- EIP-7823: caps input size for the MODEXP operation at 8192 bits.
- EIP-7825: sets a per-transaction gas cap of 2²⁴ units.
- EIP-7883: raises gas costs for large exponents in MODEXP to better match computational effort.
- EIP-7934: caps execution block size at 10 MB.
Together, these changes reduce the risk that extreme transactions or oversized blocks could overload clients, stall propagation, or create instability.
New tools for users and developers
Fusaka also aims to improve usability.
For users, EIP-7917 introduces preconfirmation support. This lets wallets and applications look ahead in the proposer schedule, enabling users to lock in that their transaction will appear in an upcoming block. The result is lower latency and less uncertainty about inclusion.
For developers, Fusaka adds two notable features:
- A CLZ opcode (count leading zeros), useful for cryptographic routines and contract efficiency.
- EIP-7951, which provides native secp256r1 (P-256) signature verification. This is a common elliptic curve used in hardware devices and mobile systems, and its addition improves compatibility and account abstraction.
These changes are meant to lower friction for application developers and pave the way for new wallet designs and security models.
What ETH holders need to know
For everyday Ethereum users, Fusaka requires no action. Account balances, tokens, and applications will continue to function as before. Ethereum.org emphasizes that users should ignore scams asking them to “upgrade” ETH or transfer funds—there is no such requirement.
The responsibility lies with validators and node operators, who must upgrade their execution and consensus clients in step. Coordination remains a delicate process: if validators fall out of sync, the network risks downtime or temporary chain splits.
Fusaka will activate on Ethereum’s Holesky testnet on October 1, followed by Sepolia on October 14, Hoodi on October 28 and the mainnet launch on December 3.
At a high-level, here’s what developers agreed should be the schedule for Fusaka moving forward:
October 1: Holesky upgrade
October 14: Sepolia upgrade
October 28: Hoodi upgrade
December 3: Mainnet upgrade— Christine D. Kim (@christine_dkim) September 18, 2025
The future of Ethereum after Fusaka
Fusaka represents one of the boldest steps in Ethereum’s roadmap since the Merge. It is an attempt to deliver more blob capacity, tighter defenses, and new developer tools in a single coordinated release.
Testing and devnet trials are underway, with client teams focusing on PeerDAS performance, blob pricing models, and compatibility across execution and consensus software. If successful, Fusaka could mark a turning point in Ethereum’s ability to scale for the next wave of layer-2 adoption.
