Vitalik Buterin presented a new long-term roadmap for Ethereum, describing the “Lean Ethereum” project as the network’s third major overhaul. The proposal was disclosed on July 4, 2026, a few days after a meeting of researchers in Berlin focused on the future of the protocol.
The initiative is drawing the attention of developers, layer 2 networks, and ETH investors, as it involves profound changes to the blockchain infrastructure while maintaining compatibility with applications already in operation.
According to Buterin, the transformation will not happen immediately. He explained that “it is a set of improvements that will be implemented on the Ethereum network over three or four years.”
The proposal positions Lean Ethereum as an evolution comparable to the Merge, when the network migrated from the proof-of-work model to proof-of-stake. This time, the scope is even broader, with the gradual replacement of central components of the protocol.
Buterin highlighted that “almost all the main parts of the protocol will be replaced,” including changes to transaction processing, which may adopt recursive verification based on STARK, in addition to the replacement of elements vulnerable to quantum computing.
The plan also provides for adjustments to the consensus mechanism to allow faster finality, new gas fee charging models, and changes to the network’s data and client architecture.
Despite the magnitude of the changes, Buterin indicated that the implementation will be carried out carefully to avoid impacts on existing applications. He recalled the experience of the Merge by stating: "We’ve done this before (the Merge), we can do it again."
Privacy appears as a priority in the new model. Buterin stated that “privacy is no longer an afterthought; it is a primary goal,” indicating that new protocol components are already being designed with this focus from the outset.
Another relevant point is the evolution of data storage. There is consensus on maintaining the current dynamic state model with moderate growth, while new more scalable formats should be introduced to support applications such as ERC-20 tokens, NFTs, and DeFi.
The projection suggests that, by 2030, Ethereum may operate with about 2 TB under the current model and up to 100 TB in new more efficient formats. Migration will be optional, but economically incentivized.
The roadmap also considers deeper changes, such as the possible replacement of the Ethereum Virtual Machine with alternatives like RISC-V or leanISA, with the EVM acting as a compilation target.
The proposals come at a time of internal adjustments at the Ethereum Foundation, which recently reduced its budget and staff. Even so, technical development continues to advance with new parallel initiatives led by former researchers from the organization.