Ethereum co-founder Vitalik Buterin has put forward a new design proposal that could significantly reduce the technical and hardware barriers to running an Ethereum node. The concept, called “partially stateless nodes,” aims to decentralize the network further by enabling everyday users to run nodes on personal devices—without needing to store the entire blockchain history.
Published in a detailed blog post on Sunday, Buterin’s proposal outlines a model that would allow node operators to store only the subset of Ethereum data most relevant to them, rather than the full 1.3 terabytes (TB) of data required for operating a full node today.
Currently, running an Ethereum node requires substantial disk space and technical expertise, making it feasible primarily for institutions or developers with access to high-performance hardware. Many users instead rely on third-party services for blockchain access, which often involves trade-offs in privacy, censorship resistance, and decentralization.
Buterin’s vision is a “local-first” approach that mirrors a library system—users store the “books” they frequently access while retrieving the rest from the network as needed.
"This type of node would give the benefits of direct local access to the state that a user needs to care about, as well as maximal full privacy of access to that state," he wrote.
How It Works
The model allows node operators to customize the data their node stores, such as frequently used smart contracts, tokens, or specific decentralized applications (dApps). This configuration would be determined via a simple on-chain setting. Instead of storing complex cryptographic structures like Merkle proofs, users would only need the raw data, making the setup less resource-intensive.
On-demand verification of other blockchain segments would be done using cryptographic proofs, ensuring that trust and validation mechanisms remain intact even without full local data storage.
Building on EIP-4444
The concept of partially stateless nodes builds on the ongoing Ethereum Improvement Proposal EIP-4444, which aims to reduce node data requirements by limiting historical block data storage to just 36 days. Older historical data would be made accessible through distributed storage using erasure coding, ensuring permanence without burdening individual nodes.
While the stateless node proposal remains in its early stages, it aligns with Ethereum’s broader decentralization roadmap. The ultimate goal is to make node participation as lightweight and private as possible, thereby boosting the number of independently run nodes and reducing reliance on centralized infrastructure providers like Infura or Alchemy.
What’s Next?
The Ethereum community is expected to discuss and refine the idea over the coming months, with any formal adoption likely requiring testnet trials and further EIP developments. However, the direction signals a renewed push toward true decentralization, making Ethereum more resilient and user-friendly.
If successfully implemented, partially stateless nodes could mark a turning point for the network, opening up node operation to a broader audience and reinforcing the decentralized ethos at Ethereum’s core.
Editorial Note: This news article was originally published in cryptoindiamagazine.com
