NEAR Protocol (NEAR), a high-performance smart contract platform, might be the first L1 blockchain to achieve 1 million transactions per second bandwidth as soon as this year. As explained by its R&D unit lead Bowen Wang, NEAR Protocol will accomplish this numbers thanks to dynamic resharding architecture.
NEAR is ready to hit 1 million TPS by end of 2025, here’s how
NEAR Protocol (NEAR), a popular programmable Layer-1 blockchain, might reach 1 million transactions per second (TPS) as soon as this year, points out Bowen Wang, founder of NEAR One, R&D unit overseeing the tech development of the NEAR blockchain.
We’re aiming for @NEARProtocol to process 1 million transactions per second within this year, starting in a testing network. The goal is to prove how NEAR’s sharding design can support the massive influx of volume our industry will likely see from AI agents
— Bowen Wang (@BowenWang18) May 30, 2025
This development, with 1 million TPS, will be first released on the testnet, Wang explains. Such developments are inevitable for mainstream blockchains to fuel the new Web3 use cases associated with artificial intelligence (AI) and machine learning (ML):
The goal is to prove how NEAR’s sharding design can support the massive influx of volume our industry will likely see from AI agents
Meanwhile, Nightshade 2.0, the current approach NEAR Protocol leverages for its sharding design, is already able to process this amount of transaction workload. Sharding refers to network architecture design when the blockchain relies on an ecosystem of interconnected sub-blockchains (shards).
At the same time, NEAR Blockchain might migrate to the “dynamic resharding” model where shards automatically split and merge according to network load.
NEAR Protocol will also change the design of its validation instruments. When zero-knowledge proving becomes fast enough, NEAR is expected to replace the merkle-proof based state witness with ZK proofs.
As covered by U.Today previously, in January 2025, NEAR Protocol kickstarted a partnership with DWF Labs, one of the most influential market-making and trading firms.
Community remains pessimistic
The aforementioned developments will significantly lower the overhead in both network bandwidth usage and verification time and pave the path for even better scalability, Bowen concluded.
He also added that this approach is the only viable way to scale modern blockchains. Such scaling should only boost the opportunities of L1 since the rollup-based L2 experiments will amplify overhead in cross-chain or cross-rollup interactions.
This, in turn, will destroy user experience and discourage new cohorts of users from using crypto.
The commentators to the thread didn’t show much enthusiasm about the prospects of the developments, being highly skeptical about the number of users on the NEAR blockchain.
