Ethereum, launched in 2015, is now in its eleventh year. After The Merge, Shapella, Dencun, Pectra and Fusaka, the network is still being reworked at a structural level. For investors and teams building on Ethereum, the long-term direction matters as much as the next upgrade.
In July 2025, one day after Ethereum’s tenth anniversary, Ethereum Foundation researcher Justin Drake introduced the idea of Lean Ethereum. The term “lean” referred to stripping out redundant design choices and pulling scattered research tracks into one unified plan. In February 2026, that vision was turned into a draft document called “Strawmap.” The paper is not final, has continued to change through community discussion, and was revised again on June 26. Its latest version maps out hard forks from the Glamsterdam upgrade in the second half of 2026 through 2029 and ties them to five “North Star” goals.
A faster L1 with finality measured in seconds
The first goal is speed at the Ethereum base layer itself. Right now, a transaction on Ethereum takes about 15 minutes to reach finality, meaning the result can no longer be reversed. One reason is the size of the validator set. The network has roughly 880,000 validators, and the system cannot gather and process signatures from all of them for every block in one shot.
Instead, validators are split into 32 groups that vote in rotation, and finality takes two full voting rounds. The Lean Ethereum approach aims to remove that bottleneck with zero-knowledge proofs. Hundreds of thousands of votes can be compressed into a short mathematical proof, allowing all validators to participate in every block while aggregating the result immediately. The end goal is single-round finality.
Block production is also set to accelerate. Ethereum currently produces a block every 12 seconds. That window is split into three 4-second segments for block proposal, validator voting and vote aggregation. Those timing assumptions were set in 2020, when the beacon chain was designed around home hardware and slower internet conditions. Client software and global network performance have improved since then, and measured latency is now well below the original limits.
Under the roadmap, block time would fall from 12 seconds to 6 seconds in 2027–2028. In 2029–2030, a leaner consensus design and a new peer-to-peer communication protocol would push propagation efficiency higher and cut block time again to 4 seconds. At that point, the average time from transaction submission to execution would be about 2 seconds. Combined with faster finality, the full path from submission to irreversible confirmation would shrink to a matter of seconds.
The same consensus overhaul also targets a lower entry threshold for validators. The minimum staking requirement would move from 32 ETH to 1 ETH, widening direct participation in consensus and governance. In the version described here, Ethereum wants to be faster by 2030 without giving up decentralization, and ideally by broadening it.
A 1 billion gas per second target for Layer 1
The second North Star goal is throughput. Gas is Ethereum’s unit for measuring computation. Today, Layer 1 handles about 5 million gas per second. The long-term target is 1 billion gas per second, or gigagas, roughly 200 times current capacity.
The main obstacle is the repeated execution model that underpins blockchains. When a new block appears, thousands of nodes around the world independently rerun every transaction in that block to verify the result. That preserves trust in a decentralized system, but it also creates a lasting trade-off. If ordinary users are meant to run nodes, throughput has to stay constrained. If throughput rises sharply, the system tends to depend on a smaller set of more powerful machines.
Here again, the proposed answer is zero-knowledge proving. The node that builds a block would also submit a proof showing that every transaction in the block was executed correctly. Other nodes would no longer need to re-execute those transactions. They would only verify the proof, a task light enough to run even on a phone. Applying that model to Layer 1 block verification is what the roadmap calls L1 ZK-EVM. The Ethereum Foundation published an official roadmap for that work in February 2026.
According to Strawmap, the transition would come in stages. First, traditional re-execution and ZK proofs would run in parallel, with proofs as an optional path. In 2028–2029, the network would move to mandatory proofs, meaning blocks would be validated through ZK proofs alone. Once that happens, the cost of validation at the node level would no longer rise in step with the number of transactions in a block. That is the foundation for raising the gas limit over time and making a run at 1 billion gas per second.
For users, the change is not just about speed. Wallets on phones and in browsers could fully verify the chain on their own rather than relying on third-party RPC providers. In the language of the article, Ethereum would move from being merely decentralized to being trustless in a more direct sense.
A 1 GB-per-second data path for a much larger L2 ecosystem
The third goal is aimed at Layer 2. L2 networks execute transactions outside the Ethereum base layer and then post final data back to L1 to inherit its security. The dedicated storage space on L1 for that data is called blob space, and the total blob capacity sets the throughput ceiling for the L2 ecosystem as a whole.
In December 2025, the Fusaka hard fork introduced PeerDAS, which lets nodes verify only part of blob data. That laid the groundwork for a major increase in blob capacity later on. Candidate proposals for Glamsterdam would push blob transmission and storage efficiency further. Strawmap sketches a gradual year-by-year expansion until Ethereum reaches 1 GB per second of data bandwidth, enough to support an L2 ecosystem at the trillion-gas scale. The article compares that capacity to sending the data volume of a high-definition video every few seconds.
The roadmap also draws a clearer line around Ethereum’s long-term division of labor. Even if all of these upgrades land, Layer 1 would still have a structural throughput ceiling relative to high-performance chains such as Solana. Applications that need very high performance are expected to connect to Ethereum through dedicated L2s rather than push all activity directly onto L1.
The piece cites Robinhood Chain as an example. Robinhood has launched its own Ethereum L2 to support tokenized stock trading, inheriting Ethereum’s security while handling business-specific needs, including compliance-related arrangements, at the L2 level.
If that model holds, Ethereum in 2030 would look more like a large settlement network: Layer 1 for final settlement and security, and multiple Layer 2 networks optimized for different use cases.
A quantum-resistant base layer
The fourth North Star goal is protection against quantum attacks. Ethereum wallets currently use ECDSA for signatures, while validators use BLS. If sufficiently powerful quantum computers arrive, both cryptographic systems could be broken.
The timeline cited in the article points to growing urgency. The U.S. National Institute of Standards and Technology, or NIST, has indicated that ECDSA should be phased out starting in 2030 and fully disallowed by 2035. Google has set 2029 as the deadline for migrating its internal systems to post-quantum cryptography. In that framing, the technology capable of threatening Ethereum’s current cryptography may be arriving faster than much of the market had expected.
The proposed answer is to replace current signature schemes with hash-based cryptography. That class of systems is presented as resistant to quantum attacks, and hash signatures also fit well with ZK proving, which links this effort to the broader high-throughput roadmap. In January 2026, the Ethereum Foundation created a dedicated post-quantum research team and set a $1 million prize to test the security of the relevant hash functions.
Strawmap says the migration would stretch across several hard forks rather than happen all at once. The sequence begins with support for pre-registering quantum-resistant public keys, followed by changes to validator vote signatures, the transaction layer and the data layer. The target is a hash-cryptography-based Layer 1 by 2029. If the plan lands as described, Ethereum aims to keep user assets secure even in a quantum computing era.
Native privacy at Layer 1
The fifth goal covers something Ethereum has never had at the base layer: native privacy. Every Ethereum transaction is public today. Anyone with an address can inspect an account balance and the full history of transfers tied to it. For individuals, that creates privacy risks. For businesses, it makes many on-chain use cases difficult, including payroll and supplier payments.
The L1 privacy plan is built around shielded transfers. The article compares today’s transactions to postcards, visible to anyone who handles them, and the future design to sealed envelopes. Sender, receiver and transaction amount would be hidden from public view, while ZK proofs would still show the network that the transaction follows the rules. That is why ZK proving appears not only in the performance roadmap but in the privacy track as well.
Under Strawmap, the earliest privacy infrastructure could arrive in the Hegotá hard fork. A later phase would redesign the transaction mempool so transaction contents are encrypted before they are included on-chain. Of the five goals, privacy remains the least defined and the most exposed to regulatory conditions, making it the most uncertain. Even so, this is the first time Ethereum has placed native Layer 1 privacy inside an official roadmap, which is a notable shift by itself.
A smaller foundation and a wider development base
The article closes by summarizing the five goals in plain terms: a faster L1 that cuts finality from about 15 minutes to seconds and block time from 12 seconds to 4 seconds; a 1 billion gas Layer 1 powered by ZK proofs instead of repeated execution; a trillion-gas-scale L2 ecosystem supported by 1 GB per second of bandwidth; a quantum-resistant L1 built on hash-based cryptography; and a privacy-native L1 that hides transaction details while preserving rule enforcement.
It also notes that the group driving these changes is shifting. The Ethereum Foundation has recently reduced staff by about 20% and is narrowing its role, while newer organizations are stepping in. In June 2026, Ethlabs was formed as a nonprofit R&D institution with support from Ethereum ecosystem companies and Ethereum co-founder Joseph Lubin. A week later, Ethereum Institutional was launched by former members of the foundation’s enterprise team. The direction described here is a more purely open-source Ethereum, with a smaller foundation working alongside a broader set of independent institutions.
Still a draft, not a fixed delivery schedule
None of that means the timeline is locked. Strawmap is still a draft under active revision, not a final roadmap. The next major step, the Glamsterdam hard fork, has already slipped from the first half of 2026 to the second half. Development work could move back again. Cryptographic changes may require longer security audits, and the feature set for each hard fork can still change as researchers and the broader community continue to debate the details.
Even with those caveats, the document makes the intended destination much clearer. Ethereum’s stated direction is to stay true to its core role while becoming faster, carrying more load and raising its security ceiling at the same time. The exact form that Ethereum takes in 2030 is still open, but the targets have now been laid out in unusually explicit terms.