Ethereum’s 11th year is being defined by two parallel shifts: changes to the protocol itself and changes to the institutions around it.
On-chain, Fusaka has already gone live on mainnet and laid groundwork for handling more Layer 2 data. Off-chain, the Ethereum Foundation has gone through a major reorganization, with parts of research, institutional outreach, and privacy-related work moving into independent entities. At the same time, Lean Ethereum and Strawmap have brought the next wave of protocol redesign into open discussion, touching consensus, verification, cryptography, and state architecture.
In 2015, Ethereum put the idea of a “world computer” on-chain. By 2026, the question is no longer how to launch that machine, but how to rebuild core parts of it while it remains live and already supports large amounts of capital, applications, and users.
The first decade in brief
Ethereum mainnet launched as Frontier on July 30, 2015, turning the white paper into a public network where anyone could deploy and run smart contracts.
The next ten years were marked by a series of milestones. In 2016, The DAO hack led to a hard fork, with the non-forked chain continuing as Ethereum Classic. In 2020, the Beacon Chain went live. In 2022, The Merge moved Ethereum from proof-of-work to proof-of-stake. In 2023, Shapella enabled staking withdrawals. In 2024, Dencun introduced blobs through EIP-4844, creating cheaper data space for rollups.
Then in May 2025, Pectra improved account functionality through EIP-7702 and raised the maximum effective balance for a single validator to 2048 ETH.
Ethereum’s path was never a clean march along a fixed blueprint. Its defining trait has been the ability to keep running while replacing core parts of itself. The first decade built the base. The past year started to redefine where the network goes next.
Fusaka: nodes no longer need every blob in full
Fusaka activated on mainnet on Dec. 3, 2025. It was Ethereum’s second major mainnet upgrade of 2025 after Pectra.
The central change in Fusaka was PeerDAS. Before that, nodes had to download full blob data to check availability. PeerDAS lets nodes verify only sampled portions of the data, while a combination of work sharing across nodes and erasure coding is used to confirm that the full dataset exists.
That changes the burden on individual nodes. Instead of requiring each one to download every blob, the network can rely on partial verification across participants and still reach a reliable conclusion about data availability.
After Fusaka, Ethereum also used two blob parameter only adjustments, or BPOs, to raise the blob target per block from 6 to 14 and the ceiling from 9 to 21. The default block gas limit was also increased from 45 million to 60 million.
Fusaka looks more like structural work beneath the surface than an endpoint for performance scaling. Users may not feel a dramatic change right away, but later gains in blob capacity and lower L2 costs depend on this layer being in place.
The Ethereum Foundation shrinks while functions move outward
The protocol was not the only thing that changed. The Ethereum Foundation also went through one of the largest organizational adjustments in its history over the past year.
In March 2026, the foundation published its Mandate, naming user sovereignty as a core objective and grouping censorship resistance, open source, privacy, and security into what it called the non-negotiable CROPS principles.
On June 23, it said a months-long restructuring had been completed. A total of 54 people left, or about 20% of the original team.
After the reorganization, the foundation set out five areas of work: protocol, access, user, community, and institutional. It also kept an operations cluster plus management and support teams. Rather than trying to keep protocol R&D, product access, ecosystem communication, and institutional outreach inside one body, it chose to focus limited resources on work it believes only the foundation can sustain over the long term.
The restructuring also followed the treasury policy the foundation had published earlier. That policy started from annual operating expenses at about 15% of the treasury in 2025 and aims to bring that figure down roughly linearly to a long-term level of 5% over five years. The plan concentrates resources on the core protocol, autonomous access, privacy, and security, while more ecosystem functions are handled by independent organizations.
Several new entities involving former foundation members have already started operating independently.
- Ethlabs is an independent non-profit R&D institution. Its team includes several former Ethereum Foundation researchers, and its work is focused on Ethereum protocol research, scaling, interoperability, and infrastructure.
- Ethereum Institutional was incubated inside the foundation and now operates as an independent non-profit. It presents itself as a neutral entry point for banks, asset managers, and public institutions entering the Ethereum ecosystem, covering education, demand gathering, and ecosystem coordination. On July 29, it said it had completed an initial ecosystem funding round and formed a supporters alliance, with backing from more than 100 ecosystem participants. The round was led by BitMine, SharpLink, and Ethereum co-founders Joseph Lubin and Mihai Alisie. No amount was disclosed.
- EthSystems continues work previously handled by the foundation’s Institutional Privacy Task Force, but as an independent for-profit company building privacy, compliance, and confidential transaction systems for institutions.
Strictly speaking, these groups should not be described in blanket terms as legal spinouts of the Ethereum Foundation. Ethlabs was created by former foundation researchers. Ethereum Institutional was incubated internally and later became independent. EthSystems took over related team functions and work. A more precise description is that the foundation is allowing part of its talent base, funding sources, and specialist functions to move outside its walls and form organizations that can raise capital and make decisions on their own.
That does not directly change how Ethereum protocol governance works, but it does change the organizational structure around protocol research, institutional engagement, and product building. Those functions no longer need to sit on the same balance sheet.
Lean Ethereum and Strawmap: a vision and a draft map
Behind the organizational changes is a longer technical line of development.
On July 31, 2025, the day after Ethereum’s 10th anniversary, Justin Drake published the Lean Ethereum vision and proposed a more aggressive path for the protocol’s next decade.
At the consensus layer, Lean Ethereum aims to compress finality to a matter of seconds. At the data layer, it points to a next-generation data availability sampling approach to expand blob capacity. At the execution layer, it proposes a leaner instruction set better suited to SNARK proofs while trying to preserve EVM compatibility and existing network effects as much as possible.
On top of those changes, the protocol would also need to replace signature schemes and other cryptographic components that could face threats from quantum computing, with long-term security in mind.
The long-range targets laid out in Lean Ethereum include roughly 1 gigagas per second on L1 and 1 teragas per second on L2, which the document estimates as about 10,000 TPS and 10 million TPS respectively. The document was explicitly labeled as Justin Drake’s personal vision when released. Its purpose was to push community discussion, not to serve as an already approved upgrade commitment.
In February 2026, Strawmap went live under the maintenance of EF Architecture. Justin Drake, Vitalik Buterin, and others have been involved in maintaining and organizing it. The name combines “strawman” and “roadmap,” and it is best understood as a draft route map still open to change by the community.
On July 4, Vitalik Buterin published an updated interpretation after the Berlin researchers’ meeting. He said the main parts of Lean Ethereum would likely take three to four years and arrive across multiple upgrades. Recursive STARKs, post-quantum cryptography, consensus and finality redesign, multidimensional gas, new state types, and client architecture were all listed as areas under discussion. He described the phase as a “third major iteration” after early Ethereum and The Merge, and said privacy had become a first-class goal.
Strawmap currently lists five directional goals: Fast L1, Gigagas L1, Teragas L2, Post-Quantum L1, and Private L1. In plain terms, that means faster confirmation, higher mainnet throughput, more L2 data space, post-quantum cryptography, and protocol-native privacy.
These remain research directions and engineering targets. Full deployment would still require substantial design, implementation, testing, and community coordination.
Four data sets that frame Ethereum at age 11
One way to understand Ethereum’s current position is to look at four sets of numbers.
Stablecoins: mainnet still carries roughly half the supply
As of July 31, 2026, RWA.xyz put stablecoins on Ethereum mainnet at about $155.9 billion. Against its global figure of about $296.9 billion, that works out to roughly 52.5%.
DeFiLlama, using a different asset coverage and network classification method, put stablecoins on Ethereum at about $146.9 billion, or around 49.0% of a total stablecoin market of roughly $300.1 billion.
The two totals differ because of token coverage, cross-chain asset handling, and methodology, but both point in the same direction: Ethereum mainnet still hosts close to half of global stablecoin supply.
Tokenized real-world assets: Ethereum mainnet leads
RWA.xyz said that as of July 31, 2026, Ethereum mainnet hosted 1,552 RWAs, with on-chain distributed asset value at about $17.15 billion. That was about 3.3 times the figure for second-ranked BNB Chain, putting Ethereum first among the 38 networks tracked in its dataset.
In that context, on-chain distributed assets refer to tokens that can move off the issuer’s platform and transfer between wallets that meet whitelist or investor eligibility requirements.
Ethereum Institutional cited a broader ecosystem view from RWA.xyz and said Ethereum plus its L2s carry more than 60% of stablecoin supply and more than 75% of tokenized RWAs. That framing includes L2s and a broader asset set, so it should not be mixed directly with RWA.xyz’s L1-only data in a straight comparison.
DeFi: Ethereum L1 still dominates by single-network TVL
According to DeFiLlama, Ethereum L1 had about $41.2 billion in DeFi TVL, accounting for nearly 55% of the total across all networks. The next two, BSC and Tron, were both near $4.9 billion. On a single-network basis, Ethereum’s TVL was about eight times that of the runner-up.
L2 activity: execution has moved outward
Data from growthepie showed that as of July 31, 2026, the Ethereum L2 networks it tracks processed about 24.61 million transactions per day in aggregate, 13.9 times Ethereum mainnet’s 1.77 million. Weekly volume was about 159.9 million, or 9.76 times mainnet’s 16.38 million. Monthly volume reached about 661.46 million, 10.4 times mainnet’s 63.51 million.
Those figures show that a large share of execution activity has moved to the scaling layer. At the same time, stablecoins, RWAs, and DeFi TVL remain heavily concentrated on L1.
Robinhood Chain offers a concrete example. Its public mainnet launched on July 1, 2026. Built on the Arbitrum stack, it is an EVM-compatible L2 that uses ETH for gas and settles to Ethereum, with a focus on stock tokens and other financial assets. A brokerage platform with a large retail user base has started placing part of its future financial infrastructure inside the Ethereum ecosystem.
None of those datasets or examples alone proves Ethereum is already a global financial settlement layer. What they do show is that Ethereum has built significant scale in stablecoins, DeFi, tokenized assets, and L2 execution, and has started drawing traditional financial platforms to build on-chain infrastructure for real users.
What comes next: Glamsterdam, then Hegotá
Ethereum’s next mainnet upgrade is Glamsterdam, a combined name taken from the execution-layer upgrade Amsterdam and the consensus-layer upgrade Gloas.
As of July 30, 2026, the official planning window was the second half of 2026, though no mainnet activation date had been published.
Glamsterdam has two core features. The first is ePBS, which writes proposer-builder separation into the Ethereum protocol, reducing reliance on off-protocol relays and leaving more time for block propagation. The second is BAL, a block-level access list that lets nodes know in advance which accounts and storage locations a block will read and modify, creating a basis for parallel execution and faster sync.
Developers said in testing that ePBS, BAL, and EIP-8037 repricing together could provide the technical basis for a post-upgrade gas limit baseline of 200 million.
After that comes Hegotá. No mainnet date has been announced, and it may land in 2027 depending to a large extent on how Glamsterdam progresses over the coming months.
Hegotá has already selected FOCIL, or EIP-7805, as its core consensus-layer feature. The mechanism uses in-protocol transaction inclusion lists so that more validators participate in deciding which transactions should make it into a block, reducing the ability of any single block builder to keep filtering transactions over time and improving censorship resistance.
Frame Transaction, or EIP-8141, is currently listed as under consideration. The idea is to separate transaction validation, execution, and gas payment into different frames, creating a base for native account abstraction, flexible signatures, and future post-quantum account migration. But under consideration does not mean confirmed for inclusion. Client teams may still revise the design or choose another approach.
Why this year stands out
The deepest changes in Ethereum over the past year have unfolded along two lines at once.
On the technical side, Lean Ethereum and Strawmap have pushed discussion toward a system-level redesign of consensus, verification, cryptography, and state structure. On the organizational side, the Ethereum Foundation has deliberately narrowed its boundary, while a share of the work once handled internally is being carried forward by independent entities.
Eleven years ago, the first Frontier block took the idea of a world computer from a white paper into production. Eleven years later, Ethereum faces a harder problem: how to rebuild a public network that already carries large amounts of assets, applications, and users without taking it offline.