Why Are More Projects Moving Back to the Ethereum Mainnet from L2?

Ethereum's scalability is rewriting the underlying logic, and the once-popular Rollup narrative is undergoing a reassessment.

Written by: William M. Peaster, Bankless
Compiled by: Saoirse, Foresight News

The Ethereum Name Service (ENS) has just completed its largest upgrade to date. Notably, this upgrade was deployed directly on the Ethereum mainnet, rather than on a Layer 2 network that had been in preparation for years.

On Tuesday, October 27, ENS v2 officially launched on L1. Nine months prior, ENS had already abandoned its original plan to host this upgrade on the Rollup chain Namechain.

ENS is not an isolated case. Over the past year, an increasing number of projects have chosen to exit Layer 2 networks, shelve plans for self-built Layer 2s, or shut down existing dedicated chains, migrating core contracts back to Ethereum L1.

In short, this is a wave of returning to the mainnet.

The underlying background of this trend: the gas limit on Ethereum (the amount of computation a single block can carry) will increase from 30 million to 60 million by 2025, leading to a significant drop in L1 transaction fees.

Ethereum will phase in an increase of the block gas limit from 30 million to 60 million by 2025, marking key events such as Synthetix's publication, Vitalik's article, and the launch of ENS v2 on L1. The Glamsterdam upgrade has already reached a gas limit of 200 million on the Sepolia testnet, but the mainnet launch date is yet to be determined.

At the same time, the cost of building a public chain has proven to be extremely high. Just this month, both Blast and Abstract announced the shutdown of their Layer 2 networks within four days.

Let’s outline the entire situation, starting with this latest domino piece.

Why ENS Chose to Stay on Ethereum

As many of you know, ENS can convert lengthy wallet addresses into easily readable and memorable domain names like "william.eth."

ENS v2 is a complete overhaul of the entire protocol. Simply put, now every ENS domain can build an independent registration system for its subdomains; domain holders will gain fine-grained permission control and can pay related fees using stablecoins like USDC.

However, the most noteworthy aspect of this article is the deployment location of v2. Back in May 2024, I wrote an introductory piece on ENS v2, discussing ENS Labs' plans to migrate the protocol to a Layer 2 network, which later materialized as a dedicated Rollup—Namechain. At that time, I believed this upgrade could usher ENS into a new phase of mainstream adoption.

ENS v2 indeed marks the beginning of a new phase, but it is not on L2. On February 6, ENS founder Nick Johnson explained this shift in direction: "Ethereum L1 is scaling, and the speed of scaling far exceeds what most people expected two years ago."

The facts have confirmed this: with Ethereum's scaling, the gas costs for registering ENS domains have dropped by 99% over the past year. Additionally, Nick pointed out that Layer 2 networks introduce additional trust assumptions that do not exist in L1 itself.

Vitalik Buterin also publicly endorsed this decision on the same day: "This is a wise decision!"

The Dominoes That Keep Falling

ENS is the latest member of this "return list," but certainly not the first.

This summarizes projects that have migrated back to the Ethereum mainnet from Layer 2 networks (or originally planned self-built Layer 2s) in 2026, marking the platforms they migrated from, the time, and the migration progress.

Synthetix was an early mover. It was the first to launch on Ethereum Rollup back in 2021, but in June 2025, it published an article titled "Layer 2 is Dead, Long Live Ethereum," pointing directly to issues like liquidity fragmentation.

Thus, Synthetix gradually phased out its L2 deployment that summer, and its new perpetual contract exchange landed on L1 in December. Founder Kain Warwick later admitted: "I spent two years pushing the L2 narrative, but this path was not viable for Synthetix, so we chose to return to the mainnet."

Then in February, Vitalik Buterin published a viewpoint: as L1 continues to scale, the process for Layer 2 networks to achieve complete decentralization is much slower and more difficult than initially envisioned. He concluded that the original vision for Layer 2 networks and their positioning within the Ethereum ecosystem is no longer valid.

Three days later, ENS halted the Namechain project.

This summer, Superseed became another typical case. Its original core mechanism relied on fee deductions from its own Layer 2 chain for user loans. However, the project officially abandoned this chain on August 15 and is currently migrating its entire ecosystem back to the Ethereum mainnet, replacing the original fee-based model with an application layer revenue strategy.

In September, the prediction market protocol Trueo behind DefiLlama announced its return to Ethereum from the Base chain. Vitalik expressed support for this plan.

October also saw a surge of similar movements:

• ShapeShift: On October 1, it migrated the rFOX staking program from Arbitrum to L1, as the vast majority of FOX tokens were already held on the mainnet.
• Blast: On October 2, it announced a gradual shutdown, as on-chain revenue could not cover operational expenses.
• Nillion: On October 5, it launched the Blacklight network on L1, with a roadmap indicating a gradual phase-out of its L2.
• Abstract: On October 6, it announced it would shut down on December 15, concluding that an independent public chain model aimed solely at ordinary users is not sustainable.

Several NFT projects within the Abstract ecosystem, such as Gigaverse's ROMs and BEARISH, have also announced their migration to L1.

Core State Returning to L1?

Behind these migrations lies a deeper trend: projects are separating state storage from computation execution.

Taking Synthetix's new perpetual contracts as an example: orders are matched off-chain, but user funds are held on Ethereum L1, and transactions are ultimately settled on L1.

Looking at Derive V3, which I analyzed earlier this month: to ensure transaction speed, order matching is done off-chain; risk assessment is validated through zero-knowledge proofs; user funds are stored in Ethereum L1 contracts.

ENS v2 also adopts this approach: the domain NFT itself is stored on L1; subdomains can leverage the CCIP-Read standard to achieve zero gas parsing on L2 or off-chain, resulting in lower usage costs. CCIP-Read allows domain names to point to data stored elsewhere.

In other words, high-value core components (asset custody, ownership confirmation, transaction settlement) are returning to L1, while computation-intensive tasks are run in the lowest-cost, fastest environments. This is fundamentally different from traditional Rollup models: traditional Rollups carry both state and computation, and users must go through cumbersome exit processes to retrieve their assets.

This compares three architectures: traditional Rollups store both state and execute computation on Layer 2; new hybrid applications place core states like funds and settlements on Ethereum L1, while complex computations are handled off-chain; L1.5 temporary Rollups allow user wallets to manage private states, with L1 only storing state commitments and validation proofs, without persistent data storage in between.

Of course, there are also some technical routes that are gradually blurring the boundaries between L1 and other environments.

For instance, the Ethereum Economic Zone (EEZ), which I introduced in March: its goal is to achieve synchronous composability between Ethereum and its Rollups, meaning contracts on different chains can interact directly within the same transaction.

EEZ does not aim to detach applications from L2 but rather to make Rollups feel like a part of L1 in terms of experience. Relevant plans are being implemented: GnosisDAO voted in August to transform Gnosis Chain from an independent L1 into an EEZ architecture Rollup.

There are also some updates and untested concepts. This week, researcher namnc proposed ephemeral Rollups (L1.5) on the Ethereum research forum, pushing the idea of separating state and computation to the extreme.

The logic of this design is: L1 only saves user state commitments and verifies all proofs; user wallets manage private states themselves; computations are all short-term tasks, and no state is retained after completion. Since L1.5 does not persistently store state, there is naturally no asset exit process.

However, this wave of return has one prerequisite: Ethereum L1 must continue to have sufficient computational capacity. If the influx of projects exceeds the speed of L1's scaling, gas fees will soar again, and the entire situation will reverse.

The important scaling upgrade Glamsterdam for Ethereum had its testnet launch plan postponed last week, with the Ethereum Foundation stating that the mainnet activation time is yet to be determined.

However, Glamsterdam has already raised the gas limit to 200 million on the testnet, more than three times the current limit of 60 million. Referring to the fee reduction effects from the last gas limit doubling, if this is implemented on the mainnet, L1's capacity will be quite considerable.

Regardless of how things evolve, the trend is clear: more and more projects are placing their core state data back on L1, with ENS being the latest member.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.