Ethereum's 2026 Upgrade Roadmap: Fusaka, Glamsterdam & What's Next
Two major Ethereum upgrades have already shipped or are landing in 2026 — Fusaka (December 2025) and Glamsterdam (mid-2026). If you hold ETH, use Layer 2s, or stake as a validator, these upgrades directly affect you. Here's exactly what's changing and why it matters.
1. Ethereum's Upgrade Naming Convention
Ethereum upgrades combine a consensus layer name with an execution layer name. Fusaka pairs the “Fulu” consensus upgrade with the “Osaka” execution upgrade. Glamsterdam will combine “Glam” (consensus) with “Amsterdam” (execution). Developers name them by referencing Devcon cities — a fun tradition that's also become a reliable naming pattern you'll keep seeing.
2. Fusaka: What Shipped in December 2025
Fusaka activated on December 3, 2025, directly following Pectra (May 2025). Where Pectra was about user experience (smart accounts via EIP-7702, larger validator balances), Fusaka is about bandwidth.
The headline features of Fusaka:
- PeerDAS (EIP-7594) — Flagship upgrade. Expands Ethereum's blob capacity from 6 blobs per block to 48 blobs per block (an 8x increase).
- Increased Block Gas Limit — Raised from 45 million to 150 million gas, a 3.3x expansion.
- EVM improvements — Multiple EIPs reducing smart contract execution costs and improving developer tooling.
The practical effect: Ethereum can serve far more rollup activity per second without raising fees. Layer 2s like Arbitrum, Base, and Optimism can post more data to Ethereum L1 at lower cost — which flows through to cheaper transactions for end users.
3. PeerDAS Explained — The Bandwidth Revolution
PeerDAS stands for Peer Data Availability Sampling. It's the first step toward Ethereum's full Danksharding vision — and a fundamental shift in how Ethereum's nodes handle data.
The Problem PeerDAS Solves
Every Ethereum node previously had to download every blob of data posted to the network. As blob volume grew post-Pectra, validators were downloading up to 750 MB of blob data per day — expensive and increasingly centralization-inducing.
How PeerDAS fixes it: Instead of each node downloading everything, nodes now only sample random small pieces of each blob and verify those samples cryptographically. The math (erasure coding) means you only need ~30% of the samples to reconstruct the full blob with high confidence. Net result: validators download ~85% less data — from 750 MB to around 112 MB per day.
⚡ Why this matters for you: Lower validator hardware requirements = more validators = more decentralization. And more blob space means cheaper L2 transactions.
4. What Fusaka Means for L2 Users
If you're primarily using Ethereum through Base, Arbitrum, zkSync, or any other rollup, Fusaka is already in your life — you just don't feel it directly as a UI change.
- Lower fees (already happening): With 8x more blob capacity, rollups compete less for blob space, keeping L2 fees structurally lower.
- More throughput: The increased gas limit means more complex transactions per block, which improves L2 batch processing.
- Path to 100,000+ TPS: Fusaka gets Ethereum's data layer to ~10,000 TPS in effective rollup throughput. Full Danksharding is the theoretical path to 100,000+ TPS.
5. Glamsterdam: What's Coming in 2026
Glamsterdam is Ethereum's next major upgrade, expected in mid-2026. Where Fusaka expanded throughput, Glamsterdam focuses on economic fairness and infrastructure decentralization.
The headline feature is enshrined proposer-builder separation (ePBS). Additional EIPs under discussion include Verkle tree migration progress, further validator UX improvements building on Pectra's EIP-7251, and L1 execution layer gas optimizations.
6. Enshrined PBS (ePBS) — Why It Matters
Right now, Ethereum uses an off-protocol system called MEV-Boost where validators outsource block-building to specialized “builders.” This works, but it's:
- Centralizing — A handful of builders produce ~90% of Ethereum blocks.
- Trust-dependent — Validators trust builders via relays operated by third parties.
- Fragile — If relay operators go offline or act maliciously, the system breaks.
ePBS bakes proposer-builder separation directly into Ethereum's consensus layer. Builders and proposers interact via the protocol itself, not through off-chain relays. This removes relay dependency, reduces validator risk from MEV manipulation, and makes block production more decentralized over time.
For everyday users, ePBS means fairer transaction ordering and a more censorship-resistant Ethereum. You can learn more about the current MEV landscape in our MEV guide.
7. Hegota: The Late-2026 Horizon
In December 2025, Ethereum developers named their post-Glamsterdam upgrade “Hegota” — slated for late 2026. Details are still being scoped, but likely candidates include:
- Stateless clients — Nodes that don't store the full Ethereum state, dramatically reducing sync times and hardware requirements.
- EIP-7935 (Execution Tickets) — An experimental economic mechanism to further decentralize block production.
- Full Danksharding progress — The next step beyond PeerDAS toward Ethereum's ultimate data availability vision.
8. Ethereum Upgrade Timeline at a Glance
9. What This Means for ETH Holders & Stakers
ETH Holders
- More rollup throughput = more activity = more ETH burned (EIP-1559)
- Fusaka's blob expansion is deflationary — blob fees burn ETH
- Each upgrade increases Ethereum's utility as the internet's settlement layer
ETH Stakers
- Fusaka reduces data download requirements for solo stakers (~85% less)
- Glamsterdam's ePBS changes block production, but validators don't need to act — clients handle it
- Pectra's EIP-7251 (2048 ETH max) lets large stakers consolidate validators
L2 Users
- Lower blob competition from Fusaka = cheaper Base, Arbitrum, zkSync transactions
- Higher gas limit = more complex L2 batches per block
- These benefits are already live as of December 2025
⚠️ This guide is for informational purposes only. It is not financial advice. Always do your own research before making investment decisions.
Frequently Asked Questions
Do I need to do anything during an Ethereum upgrade?
No. Hard forks are handled by node and client software maintainers. If you hold ETH in a wallet or on an exchange, you don't take any action.
What's the difference between Fusaka and Pectra?
Pectra (May 2025) improved user experience — smarter wallets, faster validator onboarding, larger validator balances. Fusaka (December 2025) is about infrastructure — more data capacity via PeerDAS and a higher gas limit.
Will Glamsterdam affect my ETH staking rewards?
Not directly. Enshrined PBS changes how blocks are built, but your staking APY is driven primarily by network participation rate and MEV tips — both of which should remain stable or improve with ePBS.
What is PeerDAS in simple terms?
Instead of every Ethereum node downloading all blob data, each node downloads small random samples and verifies them. The math guarantees data availability without full downloads — like checking a book is complete by reading random pages rather than every word.
When is Glamsterdam launching?
As of March 2026, Glamsterdam is in active development on devnets. A mid-2026 target is plausible based on developer communications, but Ethereum upgrades can shift. Follow the Ethereum Foundation blog and AllCoreDevs calls for the latest.