Ethereum’s L2s are not doing too well. Data from L2Beat shows that prominent L2s like Arbitrum and Base handle around 90% of the total Ethereum scaling traffic, while the smaller or newer ones struggle with low engagement.
According to data from L2Beat, which tracks about 136 projects, only about 27 projects currently record a daily average of 1.00 UOPS (User Operations per Second) or higher.
This means about 109 projects are currently recording less than 1 UOPS. So while the total scaling factor for the ecosystem is high at nearly 97x, the throughput can be attributed to a small group of highly active chains, while over 80% of the 135+ tracked projects endure negligible daily traffic (under 1 user operation per second).
Source: L2Beat
Ethereum L2s report underwhelming user activity
The Ethereum ecosystem has split into two, with L1 serving as the global vault while L2s have become the retail floor. This has affected metrics like user activity and transaction volume.
According to recent reports, L2s are lagging behind on total value locked and daily user activity. Ethereum currently has around $68 billion in TVL, while all its L2s combined have around 50 billion in TVL.
The daily users are also split between the top L2s like Arbitrum and Base. So while the top L2s are attracting most of the liquidity and users, the newer or less popular ones keep facing low activity.
Base, especially, has emerged as a consumer-friendly hub and often handles more daily users than the L1 itself. The biggest reason for this is that the mainnet is once again attracting users because the fee structure is now vastly different.
That difference is thanks to the Dencun and subsequent Pectra/Fusaka upgrades, which fundamentally changed the fee relationship, making things far cheaper on the mainnet.
Of course, Ethereum L2s are not completely beat, and the most dramatic divergence can be witnessed in transaction throughput, with L2s now processing millions more transactions per day than Ethereum.
According to L2Beat, the ecosystem scaling factor has also reached record highs with L2s handling over 20,000 TPS during bursts on some days while L1 remains steady at a structural limit.
What Vitalik Buterin thinks of the recent split
The current performance of L2s on Ethereum has not gone unnoticed by its founder, Vitalik Buterin.
As far as he is concerned, the “original vision of L2s and their role in Ethereum no longer makes sense, and we need a new path.”
“L1 does not need L2s to be ‘branded shards’, because L1 is itself scaling” he wrote on X. “And L2s are not able or willing to satisfy the properties that a true ‘branded shard’ would require.”
Vitalik admits that Ethereum itself is now scaling directly on L1, with large planned increases to its gas limit this year and the years ahead. He believes the natural step is to stop treating L2s as “branded shards” of Ethereum,” but instead as a full spectrum.
In his post, he also outlined what could come next for L2s that want to stand out or remain relevant, including refocusing on adding value and maintaining higher standards than L1s or supporting maximum interoperability with Ethereum.
“It’s each L2’s choice exactly what they want to build. Don’t just ‘extend L1’, figure out something new to add,” Vitalik wrote.
How major L2s responded to Vitalik’s rhetoric
Vitalik’s talk about how the rollup-centric vision of L2s no longer fits has since gone viral among crypto circles, and leaders of major L2s have shared their own opinions in response.
Steven Goldfeder, the cofounder of Offchain Labs, which is behind Arbitrum, responded with a lengthy thread where he agreed with parts of Buterin’s assessment while pushing back on downplaying scaling.
According to him, even with higher gas limits, the Ethereum mainnet can not realistically handle thousands of TPS during peak times without compromising on decentralization or costs.
Karl Floersch, Optimism’s cofounder, supports viewing L2s as a full spectrum but emphasized the need for modular designs. Floersch agreed that L2s need to go beyond being cheaper Ethereum clones and innovate to retain their place or become obscure.
He also seems to be treating the discourse as a challenge for Optimism, one he claims the network is already closer to achieving in reality.
Base’s Jesse Pollak echoed the sentiment, admitting that L1 scaling is a positive for the whole ecosystem and that L2s need to show off more unique features that can help them stand out. He claims that Base is focusing on those differences to stay relevant, which aligns with Buterin’s suggestions.
Zksync’s Alex Glukhov agreed explicitly with Buterin, claiming that L2s that want to be valuable in the future must learn to “specialize.” Meanwhile, StarkWare’s Eli Ben-Sasson has hinted that ZK-native L2s like Starknet are already on that specialization path Buterin is describing.
