Ethereum’s next major upgrade is called “Fusaka”, which combines “Fulu” (consensus) and “OSAKA” (execution), and will change how the network handles data and fees without changing the core user experience.
Beneath the surface, this is a statement of direction. While Ethereum's main chain remains the hub for final payments and data availability, day-to-day activities continue to flow outward to cheaper and faster rollups.
The open question of whether Fusaka will bring users back to layer 1 has already been answered. That won't happen. This makes it even more difficult to move away from Layer 2.
Inside the fusaka: Scaling the piping, smoothing the run
Fusaka's technical backbone is centered around data availability, sampling, and blob management. This is Ethereum's approach to making layer 2 posts cheaper and more efficient. The headline proposal, EIP-7594 (PeerDAS), allows nodes to sample only pieces of rolled-up data called “blobs” rather than downloading everything.
This frees up higher blob capacity and significantly reduces bandwidth costs for validators. This is a prerequisite for scaling L2 throughput.
Next came EIP-7892, which introduced the “Blob Parameter-Only” fork (BPO). This is a mechanism to incrementally increase the number of blobs per block (for example, from 10 to 14, or from 15 to 21) without rewriting the protocol.
This will allow developers to effectively adjust Ethereum's data capacity without waiting for a full upgrade. EIP-7918 sets a floor on the base price for blobs, ensuring that the auction price for data space does not drop to near zero when demand is low.
The rest of the bundle focuses on user experience and safety. EIP-7951 adds support for secp256r1, the cryptographic curve used by WebAuthn, allowing passkey logins between Ethereum wallets. EIP-7917 introduces deterministic proposer lookahead. This is a small but important change that helps the pre-confirmation system predict who will produce the next block, allowing for faster transaction guarantees.
Meanwhile, EIP-7825 limits transaction gas to prevent denial-of-service risks, and EIP-7935 adjusts default block gas targets to maintain validator stability.
These upgrades are already live on testnets such as Holesky and Sepolia, with mainnet activation scheduled for early December.
Why fusaka is important in fees and the roll-up economy
Fusaka is not promising cheaper layer 1 gas to users. Built to lower Layer 2 charges. This upgrade improves the economics of networks like Arbitrum, Optimism, Base, and zkSync by allowing rollups to post more data at a lower cost.
Internal modeling suggests that rollup fees can drop between 15% and 40% under normal conditions, and up to 60% if the supply of blobs exceeds demand for an extended period of time. On the Ethereum mainnet, gas prices are likely to remain roughly flat, although future adjustments to block gas targets could bring them down another 10-20%.
However, updates to passkeys and proposers could change the experience of using Ethereum. WebAuthn support allows wallets to integrate biometric or device-based logins, removing the hassle of seed phrases and passwords. Predictable proposer schedules enable proactive confirmation, so users can expect near-instant confirmation of routine transactions, especially rollups.
The end result is that Ethereum can be used more smoothly without anyone being pulled back to L1. The rails will be faster, but will still point in the direction of the roll-up lane.
L1 is payment, L2 is experience
Ethereum’s architecture is no longer a monolithic vs. modular design debate, but modular by choice. The purpose of Layer 1 is to serve as a base for high-security payments and data availability, while the actual user activity is moved to Layer 2.
Fusaka reinforces this division. Increasing blob capacity allows L2 to handle higher throughput for games, social apps, and microtransactions that would be uneconomical on mainnet. Improved login and confirmation workflows make these L2 environments feel native and immediate, eliminating many of the UX gaps that once favored L1s.
Where would users choose Layer 1? Narrow cases include high-value payments, institutional-scale transfers, or situations where block order accuracy is important, such as miner extractable value (MEV) management or DeFi liquidation. However, these scenarios are only a small part of the overall on-chain activity. For the rest, L2 remains its natural home.
The Bigger Story: Ethereum as a Layered Internet
From the top, Fusaka emphasizes maturity over gas optimization. This provides Ethereum with a scalable framework to adjust data capacity (BPO) without disruptive forks and a UX layer that brings Web3 closer to Web2.
But the philosophy is clear. The network is not trying to concentrate traffic on the mainnet. Rollup is building a highway system to handle regional traffic, and L1 will eventually serve as the courthouse where everything is notarized.
There is also a financial element to this story. Cheaper data posting could bring a wave of new low-value applications such as social, payments, and gaming back into the rollup. Each of these still consumes ETH through blob fees, and due to the fee floor in EIP-7918, these fees contribute to ETH burn. Despite lower user costs, Ethereum's burn rate could rise further if activity grows faster than fees decline.
On the validator side, PeerDAS reduces the bandwidth load, but may create new dependencies on “supernodes” that store complete BLOB data. This is a decentralization trade-off, and the community will continue to discuss how to scale data availability without narrowing participants.
The balance Ethereum strikes here between throughput, ease of use, and reliability reflects the broader direction of crypto infrastructure. L1 is hardened as a secure foundation, while L2 absorbs experimentation and scale.
Take-out
Fusaka isn't trying to regain attention on the Ethereum mainnet. Quite the opposite, it's a deliberate move to strengthen our foundation for a rollup-centric future.
This upgrade expands data capacity, stabilizes prices, and modernizes the wallet experience, but it is done to serve the layers above. Ethereum's L1 is now more secure and smarter, but users continue to live on L2, which is now cheaper and faster than before.
By the time BPO1 and BPO2 are rolled out early next year, the real signals to watch will be blob utilization and capacity, L2 fee compression, and the adoption of passkeys in wallets. The result will define how frictionless Ethereum will become in 2026, not by drawing people back to the main chain, but by making the exit nearly invisible.
