Blockchain becomes a burden: Why crypto projects are abandoning L2 networks

​Just a few years ago, launching a proprietary Layer 2 blockchain was considered an almost essential step for any major crypto project. But in 2026, several prominent L2 networks have shut down or abandoned their blockchains. Why are projects no longer willing to pay for their networks, and what does this reveal about the future of the market?

In early October, Blast became one of the most prominent examples of this crisis. Before its mainnet launch, users deposited hundreds of millions of dollars into the project, attracted by yields and expectations of an airdrop. Two and a half years later, only about $32 million remained, and the project was officially declared "dead."

View post on X Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

Four days later, Abstract, a consumer-focused L2 developed by the parent company of Pudgy Penguins, announced its closure. Loopring, ZERϴ Network, and Botanix had already shut down operations. Yet the L2 market itself has not disappeared. 

According to L2BEAT, approximately $41.7 billion is secured across Layer 2 networks, but Base and Arbitrum alone account for roughly 63% of that amount.

From $2 billion to network shutdown

Blast launched in late 2023 with a native yield model: ETH and stablecoins held within the network automatically generated returns for their owners. The project also secured $20 million from Paradigm and Standard Crypto, quickly becoming one of the most talked-about L2 networks of the new market cycle.

Much of the demand was driven by points and the prospect of an airdrop. Following the BLAST token distribution, activity began to decline, and on October 2, 2026, the team announced the network's shutdown. Developers explicitly cited the project's economics: operating costs exceeded revenue, and they saw no realistic path to a sustainable business model.

Abstract faced a similar problem, although the network could hardly be described as empty. During its operation, it attracted more than 400,000 users and processed over 325 million transactions. Nevertheless, Igloo funded the L2 for approximately a year and a half and, according to CEO Luca Netz, lost "tens of millions of dollars" on it.

Growth stalled, liquidity remained limited, and institutional demand failed to materialize. The company could have extended the network's life by issuing a token and raising additional capital, but rejected that option. Abstract will cease operations on December 15.

Technological leadership offered no protection either. Loopring, one of the pioneers of ZK-rollups, shut down its DEX and L2 operations on June 28 after the team acknowledged a lack of meaningful demand.

However, this wave of closures does not mean that scaling Ethereum has become unnecessary.

Base currently secures approximately $15.5 billion in value, while Arbitrum accounts for another $11 billion. Instead, capital and users appear to be concentrating in networks that already offer liquidity, applications, and access to a broad audience.

Networks become an operating expense

Some teams decided in 2026 not to wait until maintaining their networks became too expensive. Sophon calculated the cost of operating its L2 and reached a different conclusion: shut down the network and migrate its products to Base.

The developers were spending approximately $3.4 million annually on infrastructure, rollup services, data, and related tools. Following the migration, the team expects to save roughly $3 million per year and redirect those funds toward application development.

View post on X Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

Sophon's decision is particularly revealing because the project had not lost the ability to maintain its network. It simply concluded that owning a blockchain offered benefits that did not justify the cost. Instead of separate infrastructure, the team chose an established L2 with existing users and liquidity.

Zerion reached a similar conclusion. In May, the company shut down ZERϴ Network and redirected resources toward its core products: its wallet and API.

Botanix took a different approach. The project spent nearly four years developing an EVM-compatible Bitcoin L2 and deliberately avoided token incentives to measure genuine demand rather than activity driven by rewards.

During its operation, the network processed more than 26 million transactions and attracted over 176,000 addresses. However, those figures failed to translate into sustainable demand, and developers announced the project's closure in June.

Infrastructure providers also recognized the problem. In May, Syndicate Labs, a developer of rollup and sequencer solutions, shut down. The team said the market had contracted sharply, with several projects quietly ceasing operations for every new network launched.

The understanding of what L2 networks should offer is also changing. In 2026, Vitalik Buterin said the original model, in which Layer 2 solutions were primarily intended to scale the underlying blockchain, no longer reflected current conditions.

Ethereum's base layer is increasing throughput faster than expected, meaning that offering "the same ecosystem, only cheaper" is no longer enough.

View post on X Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

The Ethereum Foundation later outlined a similar approach: L2 networks should offer their specialization, such as privacy, higher performance or unique use cases. Otherwise, a new network must compete with Base, Arbitrum and the base layer for the same users without a clear advantage.

Lightning advances but remains a niche solution

Bitcoin faces a different version of the problem. It does not have as many competing Layer 2 networks as Ethereum, yet its best-known scaling technology has existed for years without becoming an obvious standard for payments.

Major exchanges and wallets continue to support Lightning. In 2026, Square also began automatically enabling eligible U.S. merchants to accept Bitcoin through the network. Merchants can receive payments directly in dollars, with fees set at 0% through the end of the year.

Lightning Labs continues to simplify the infrastructure as well. During the summer, it introduced Wavelength, a toolkit that hides much of the technical complexity behind an API.

View post on X Loading...

The tweet was deleted by the author.

But we saved everything 🙂.

However, technological development continues to outpace actual adoption.

According to CoinGate, the service processed 782,403 cryptocurrency payments in the first half of 2026. Bitcoin accounted for 21% of transactions, but only 9.6% of Bitcoin payments went through Lightning.

Bitcoin also lost the top position in the service's statistics to USDC, which accounted for 22.1% of all transactions. This gives USDC an important advantage in payments.

Lightning makes Bitcoin transfers cheaper and faster, but it does not eliminate the asset's price volatility. A stablecoin, by contrast, combines low-cost settlement with a predictable unit of account.

Technical barriers also remain. Lightning requires liquidity in payment channels, proper liquidity distribution, and transaction routing. The more effectively a service hides this complexity from users, the more responsibility falls on the wallet, exchange, or payment provider.

The project's history illustrates another side of the same problem: even a technically mature network does not achieve mass adoption simply because it makes transactions cheaper and faster.

Ultimately, users do not need an L2 network itself. They require a specific product: a payment, exchange, game, loan, or money transfer. That is why the developments of 2026 look less like the end of Layer 2 networks and more like the end of launching them simply for the sake of having one.

A proprietary Layer 2 blockchain must now explain not only how it works, but why it needs to exist at all.