Ethereum is moving faster than at any point since The Merge. In the span of a year it has shipped a major scaling upgrade, ramped up its data capacity in live production, and entered the final stretch on its most ambitious protocol overhaul in years. If you've been trying to keep up with what's actually changing under the hood, this guide breaks down Ethereum's most recent technical progress — Fusaka, the blob-capacity ramp, the upcoming Glamsterdam upgrade, and what comes after — in plain English.
For a foundational primer on how the network works, Phemex's Academy explainer on Ethereum is a solid starting point. This article focuses on what's new.
Quick Snapshot: Where Ethereum Stands in 2026
Ethereum remains the dominant smart-contract platform, securing the majority of DeFi, stablecoins, and tokenized real-world assets. Its 2026 roadmap is organized around one central goal: scale the network and its Layer-2 (L2) ecosystem without sacrificing decentralization. The last twelve months delivered real, shipped progress on exactly that — not just whitepapers.
Here's what changed.
The Fusaka Upgrade: Ethereum's Biggest Recent Ship
Fusaka activated on Ethereum mainnet on December 3, 2025 (epoch 411392, block 23,935,694), the second major Ethereum upgrade of that year. Its headline feature is a technology called PeerDAS (Peer Data Availability Sampling) — and it's a genuinely big deal for scaling. Fusaka also raised the default block gas limit to 60 million, lifting base-layer throughput.
What PeerDAS actually does
Ethereum's L2 rollups post their compressed transaction data to the mainnet as "blobs." Before Fusaka, every node had to download and store every blob in full, which capped the network at just three blobs per block. That storage requirement was a hard ceiling on how cheaply L2s could operate.
PeerDAS rewrites the rule. Instead of forcing every node to store all blob data, nodes now sample small pieces of it and cryptographically verify availability. The result:
- Validators download roughly 85% less blob data than before.
- Blob capacity can scale up to 8x its previous size.
- L2 transaction costs fall further, improving user experience across the ecosystem.
Blob-Parameter-Only forks
Fusaka also introduced Blob-Parameter-Only (BPO) forks (EIP-7892), a clever mechanism that lets Ethereum raise blob limits without a full hard fork each time. This means the network can scale its data capacity incrementally and safely, monitoring node health at each step.
⚡ Buy Ethereum on Phemex today → Buy ETH
The Blob Capacity Ramp: Scaling in Real Time
What makes 2026 different is that Ethereum isn't just promising scale — it's dialing it up live. Following Fusaka:
- By January 2026, the blob target was raised to 14 (from 10) and the maximum lifted to 21 (from 15).
- The plan is to keep doubling blob counts every few weeks, working toward a ceiling of 48 blobs, while developers watch closely for any adverse effects on the nodes running the network.
For end users, more blob space translates directly into cheaper L2 fees — the transactions you actually pay for on rollups get less expensive as capacity expands. It's one of the clearest examples of Ethereum's "scale through L2s" thesis playing out in production.
Glamsterdam: The Next Major Upgrade
If Fusaka was about data availability, Glamsterdam is about rebuilding how Ethereum creates and verifies blocks. Developers are describing it as one of the network's biggest changes since The Merge.
The name is a mashup of its two halves: "Amsterdam" (the execution-layer upgrade) and "Gloas" (the consensus-layer upgrade, named after a star). Phemex's blog has a dedicated breakdown in Ethereum's Glamsterdam Upgrade Explained if you want the trader-focused angle.
Timeline: hardening phase, targeting H2 2026
Originally floated for mid-2026, Glamsterdam has proven trickier and slower to build than Fusaka. As of August 2026 it sits in the hardening / testnet-preparation phase, having completed its final multi-client devnets. The internal aspirational target is a mainnet activation around the end of August 2026, but given that recent forks needed two to four months of public-testnet seasoning, a firmer base case is a September–December 2026 window. Core developers have been explicit: getting ePBS right outranks hitting any fixed date, so a further slip is possible.
The headliner features
- ePBS (Enshrined Proposer-Builder Separation, EIP-7732). This bakes proposer-builder separation directly into Ethereum's consensus rules. A builder commits to an execution payload and payment; the proposer simply chooses the bid. It's designed to improve MEV fairness, censorship resistance, and validator economics — but its implementation complexity is the primary bottleneck to testnet readiness.
- Block-Level Access Lists (EIP-7928). Blocks will declare in advance which accounts and contract data they intend to touch, letting clients preload information and moving Ethereum toward parallel execution — faster, more predictable blocks.
- Sweeping gas repricings. Glamsterdam includes a major reshaping of gas costs: high-level compute gets cheaper while state access gets more expensive. This meaningfully alters the economics of building and transacting on Ethereum.
Development status
Glamsterdam entered its final devnet phase around June 16, 2026 — engineers at the Ethereum Foundation call it "probably the largest fork since the Merge." The Soldøgn interop devnet concluded around May 2, 2026, followed by multi-client glamsterdam-devnet-5 and devnet-6 running the full slate of planned EIPs. The network is now in the last phase before public testnets and mainnet.
🚀 Trade ETH on Phemex now → ETH Price & Trade
Beyond Glamsterdam: What Hegotá Brings
Ethereum's roadmap doesn't stop there. The upgrade after Glamsterdam — codenamed Hegotá (a mashup of the "Bogota" execution-layer update and the "Heze" consensus-layer update) — is shaping up to be another structurally significant milestone. Initially eyed for H2 2026, it may realistically slip to Q4 2026–Q1 2027. Key points from the Ethereum Foundation's planning:
- Headliner: FOCIL (Fork-Choice enforced Inclusion Lists, EIP-7805) — a censorship-resistance mechanism ensuring transactions can't be quietly excluded.
- Account abstraction is committed as part of the minor feature set, a long-awaited change that would make wallets far more flexible and user-friendly.
- Verkle Trees and 2D PeerDAS remain under active discussion as future headliner candidates — both aimed at statelessness and even greater scale.
Taken together, the trajectory is clear: Ethereum is methodically working toward higher throughput, lower fees, stronger censorship resistance, and a smoother user experience.
Ongoing Work: History Expiry & Node Storage
Not every improvement is a headline hard fork. In parallel, core developers are steadily tackling a quieter but important problem: node storage pressure. As Ethereum's history grows, requiring every node to retain all historical data indefinitely raises the hardware bar for running the network — which works against decentralization.
A concrete example currently under discussion is EIP-12188, proposed by developer kevaundray, which suggests shortening the consensus-layer block retention window to reduce how much historical data nodes must store. The idea: as Ethereum's historical-data needs evolve, trimming consensus-layer block retention can optimize node resource usage without compromising the network.
The proposal is in the public review stage and has not been merged. It has drawn early support from client developers — DApplion backed the idea, and Lighthouse contributor michaelsproul noted the change wouldn't meaningfully affect Lighthouse's operation. Discussion has flagged one trade-off: pruning execution-layer history could limit some long-running nodes' ability to serve old block data. Developers involved, however, argue this poses no threat to network security or normal node operation. As with any EIP at this stage, it still requires further review and community consensus before it could ship. (Status as of August 2026.)
It's a useful reminder that a large share of Ethereum's engineering effort goes into sustainable decentralization — keeping node requirements low enough that running Ethereum stays accessible.
The Long-Term Vision: "Hyperscaling" with UTXO-Style State
Where does all of this ultimately lead? In August 2026, Vitalik Buterin publicly sketched the destination. Weighing in on community discussion, he described Ethereum's current proposed scaling strategy as aiming to give the network "the best of UTXO-style state, dynamic state, and everything in between" — so that the great majority of Ethereum's activity can be "hyperscaled" without sacrificing decentralization, ease of node running, or censorship resistance.
The technical threads behind this vision include STARK-aggregated mempools (using succinct zero-knowledge proofs to compress large batches of transactions) and UTXO-style state proposals, with researchers even floating the idea of recursively STARK-aggregating UTXO transactions at the mempool level. Notably, Buterin credited Bitcoin's Utreexo — a UTXO-set accumulator — as prior art, a rare cross-chain nod acknowledging that Bitcoiners pioneered many of these state-compression ideas.
Two things to keep in mind. First, this is a research direction and long-term vision, not a scheduled upgrade — it won't ship in Glamsterdam or Hegotá, and specifics will evolve. Second, it clarifies the "why" behind the near-term forks: Fusaka and Glamsterdam are concrete steps toward a network that can scale massively while keeping nodes light enough for anyone to run. The endgame isn't just higher throughput — it's throughput without centralization.
Why These Upgrades Matter
Zoom out and the recent technical progress ladders up to a few concrete outcomes:
- Cheaper L2 transactions. PeerDAS and the blob ramp directly lower the cost of using rollups — where most everyday Ethereum activity now happens.
- A more scalable Layer 1. Glamsterdam's block-processing overhaul and the 200M gas floor expand base-layer capacity.
- Better economics and security. ePBS and FOCIL harden validator incentives and censorship resistance.
- Institutional readiness. Predictable, efficient execution matters to the funds and asset managers increasingly building on and allocating to Ethereum — including the spot and staking ETF products that have drawn institutional flows in 2026.
For a deeper look at how these fundamentals intersect with price, ETF flows, and staking, Phemex's 2026 ETH price analysis blog is worth a read.
What It Means for ETH Holders and Traders
Technical upgrades aren't just engineering trivia — they shape Ethereum's long-term value proposition. Successful scaling keeps developers and users on the network, sustains fee revenue, and reinforces the ETH-as-productive-asset thesis (especially with staking).
That said, upgrades also carry execution risk: delays (like Glamsterdam's slip to Q3) and complex features (like ePBS) can push timelines. And ETH's price remains driven by far more than upgrades alone — macro conditions, ETF flows, and broader risk appetite all play major roles. Always track the live market before acting; you can follow the current Ethereum price on Phemex for real-time data.
Disclaimer: This article is for informational purposes only and does not constitute financial advice. Cryptocurrency markets are volatile — always do your own research (DYOR) before making investment decisions.
How to Get Started With Ethereum on Phemex
If following Ethereum's progress has you wanting exposure, getting started is straightforward:
- Learn the basics. Start with the Phemex Academy guide to Ethereum to understand smart contracts, staking, and L2s.
- Buy ETH. Follow the step-by-step How to Buy Ethereum guide — Phemex supports flexible payment methods and credits ETH to a high-security wallet.
- Trade with confidence. New to the platform? The Phemex Help Center walks you through placing your first spot, margin, or futures trade.
Phemex offers spot, margin, and futures markets for ETH, deep liquidity, and a trading engine built for speed — everything you need to act on the Ethereum thesis in one place.
FAQ
What is the latest Ethereum upgrade? Fusaka, which activated on December 3, 2025, introduced PeerDAS and expanded blob capacity. The next major upgrade, Glamsterdam, is in its testnet/hardening phase as of August 2026, with mainnet targeted for H2 2026 (aspirationally late August, realistically September–December).
What does PeerDAS do? It lets nodes sample small pieces of blob data instead of storing all of it, cutting validator data loads by ~85% and enabling up to 8x more blob capacity — which lowers Layer-2 fees.
What is Glamsterdam? Ethereum's next major upgrade (targeted H2 2026), reworking how blocks are built and verified via ePBS (EIP-7732), Block-Level Access Lists (EIP-7928), and gas repricings. It's considered the biggest change since The Merge.
Where can I buy or trade Ethereum? You can buy and trade ETH on Phemex across spot, margin, and futures. See the How to Buy Ethereum guide to get started.
The Bottom Line
Ethereum's recent technical progress is substantial and, crucially, shipping. Fusaka delivered PeerDAS and a live blob-capacity ramp that's already lowering L2 costs; Glamsterdam is set to overhaul block production in Q3 2026; and Hegotá promises account abstraction and stronger censorship resistance beyond that. The through-line is a network methodically scaling toward cheaper, faster, more robust infrastructure.
For builders, users, and investors alike, that steady execution is the real story — far more durable than any single price move. Learn the fundamentals, track the roadmap, and manage risk accordingly.
