Ethereum’s 2026 roadmap facilities on two tracks: increasing rollup knowledge capability by way of blobs whereas pushing base-layer execution increased by way of fuel restrict adjustments.
These fuel restrict adjustments rely upon validators transferring from re-executing blocks to verifying ZK execution proofs.
The primary observe is already anchored by Fusaka, which shipped Dec. 3, 2025.
Fusaka
Fusaka units up PeerDAS plus blob parameter solely (BPO) adjustments that may increase blob throughput in measured steps, in response to ethereum.org.
The second observe is much less mechanized as a result of it depends on draft EIPs, consumer implementation, and validator operations which have to remain inside decentralization constraints, together with bandwidth, block propagation, and proving market construction.
PeerDAS is positioned because the clearest “capability ramp” lever as a result of it’s designed to scale rollup knowledge availability with out forcing each node to obtain each blob.
In response to ethereum.org, blob targets don’t soar instantly at activation, then can double each few weeks as much as a most goal of 48 as builders monitor community well being.
Optimism’s group framed the upper-end case as “no less than 48 blob goal per block,” paired with a rollup-side throughput transfer from about 220 to about 3,500 UOPS underneath that concentrate on, in response to optimism.io.
Even in that framing, the sensible query for 2026 is whether or not demand arrives as blob utilization moderately than bidding up L1 execution.
One other open query is whether or not p2p stability and node bandwidth stay inside operator tolerances as BPO will increase rollout.
On the execution facet, Ethereum is already testing increased throughput by way of coordination moderately than a tough fork.
GasLimit.pics reported a contemporary fuel restrict of 60,000,000, with an about 59,990,755 24-hour common on the time proven.
That degree issues as a result of it offers a reference level for what validators have accepted in follow.
It additionally exposes the ceiling of “social scaling” earlier than latency, validation load, and mempool and MEV pipeline pressure turn into binding.
A easy method to translate fuel restrict speak into throughput ranges is fuel per second, utilizing Ethereum’s 12-second slot time (fuel per second equals fuel restrict divided by 12).
The numbers beneath maintain the mathematics express and separate base-layer EVM transactions from rollup throughput claims.
| State of affairs | Fuel restrict | Fuel/sec (≈ fuel/12) | Tx/sec at 21k fuel | Tx/sec at 120k fuel |
|---|---|---|---|---|
| Present coordination degree | 60,000,000 | 5,000,000 | ≈238 | ≈42 |
| 2× fuel restrict case | 120,000,000 | 10,000,000 | ≈476 | ≈83 |
| Excessive-end case (requires validation change) | 200,000,000 | 16,666,667 | ≈793 | ≈139 |
Glamsterdam
The deliberate 2026 improve branding wraps a number of execution-oriented concepts into “Glamsterdam,” a shorthand slate that has been mentioned round enshrined proposer-builder separation (ePBS, EIP-7732), Block-Stage Entry Lists (BALs, EIP-7928), and common repricing (EIP-7904).
Every stays in draft kind, in response to the EIP pages for EIP-7732, EIP-7928, and EIP-7904.
Repricing targets fuel schedule mismatches which have endured for years.
It argues that correcting mispriced compute can increase usable throughput whereas acknowledging DoS threat and the fact of contracts that hardcode fuel assumptions, in response to EIP-7904.
BALs are framed as plumbing for parallelism.
The EIP cites parallel disk reads, parallel transaction validation, parallel state-root computation, and “executionless state updates,” whereas estimating about 70 to 72 KiB common compressed BAL measurement as overhead, in response to EIP-7928.
In follow, these good points solely materialize if shoppers undertake concurrency throughout the actual bottlenecks.
Additionally they rely upon whether or not the additional knowledge and verification steps keep away from changing into their very own latency tax.
ePBS sits on the heart of each MEV and throughput discussions as a result of it goals to decouple execution validation from consensus validation in time, in response to EIP-7732.
That temporal slack can be the place new failure modes can present up.
An instructional paper on the “free choice drawback” for ePBS estimates choice train at about 0.82% of blocks on common underneath an 8-second choice window, reaching about 6% on high-volatility days in its modeled circumstances, in response to arXiv.
Ethereum in 2026
For 2026 planning, that analysis pushes consideration towards liveness underneath stress, not solely steady-state payment outcomes.
The extra structural wager behind “very excessive” fuel limits is validator ZK-proof adoption.
The Ethereum Basis’s “Realtime Proving” roadmap describes a staged path the place a small set of validators first runs ZK shoppers in manufacturing.
Then, solely after a supermajority of stake is snug, fuel limits can rise to ranges the place proof verification replaces re-execution for sensible validation on cheap {hardware}, in response to the muse’s July 10, 2025 put up on weblog.ethereum.org.
The identical put up lays out constraints that matter for feasibility moderately than narrative, together with concentrating on 128-bit safety (with 100-bit accepted briefly), proof measurement underneath 300 KiB, and avoiding reliance on recursive wrappers with trusted setups, in response to weblog.ethereum.org.
The scaling implication is tied to proving markets: real-time proof provide must be low-cost and credible with out concentrating right into a slender prover set that recreates right now’s relay-style dependencies in one other layer of the stack.
After Glamsterdam, “Hegota” is positioned as a later-2026 named slot that’s nonetheless about course of greater than scope.
The Ethereum Basis revealed a headliner timeline with a Jan. 8 to Feb. 4 proposal window, adopted by Feb. 5 to Feb. 26 dialogue and finalization, then a window for non-headliners, in response to weblog.ethereum.org.
A Hegotá meta-EIP exists as draft (EIP-8081) and lists gadgets as thought of moderately than locked, together with FOCIL (EIP-7805) as at present thought of, in response to EIP-8081.
The near-term reporting worth in that schedule is that it creates dated resolution factors buyers and builders can observe with out inferring commitments from codenames.
The primary is that Hegota headliner proposals shut Feb. 4.
