Ethereum’s 2026 roadmap is the most ambitious scaling push we’ve seen. With Glamsterdam and Hegota upgrades planned, plus aggressive blob expansion, we’re looking at a potential 100,000+ TPS across the L2 ecosystem. Let me break down what’s coming.
The Big Upgrades
Glamsterdam (Expected Mid-2026)
The biggest planned upgrade introduces two major changes:
1. Block Access Lists
Block producers will provide a map showing which transactions touch which parts of the system. This allows clients to run many transactions simultaneously using multiple CPU cores.
Impact:
- Faster block processing
- Better node performance
- More efficient parallel execution
2. Enshrined Proposer-Builder Separation (ePBS)
Today, block building relies on external systems like MEV-Boost. ePBS brings this directly into Ethereum’s consensus layer.
Impact:
- More decentralized block production
- Reduced MEV centralization risks
- Better protocol-level guarantees
Hegota (Expected Late 2026)
Verkle Trees
Replacing Ethereum’s current Merkle Patricia Trie with a more efficient proof system.
Impact:
- Lower hardware requirements for nodes
- Broader node participation
- More decentralized network
FOCIL (Fork-Choice Inclusion Lists)
Enhances censorship resistance by ensuring proposers can’t selectively exclude transactions.
The Blob Expansion
This is the game-changer for L2s. The number of data blobs per block is expected to rise sharply:
| Metric | Current | 2026 Target |
|---|---|---|
| Blob target per block | ~6 | 48+ |
| L2 throughput (UOPS) | ~220 | ~3,500 |
| Total ecosystem TPS | ~1,000 | 100,000+ |
Optimism’s team framed the upper-end case as rollups moving from ~220 to ~3,500 user operations per second. Multiply that across the ecosystem, and we’re talking about transaction throughput that rivals credit card networks.
Which L2s Are Positioned Best?
Based on developer activity and infrastructure readiness:
1. Base
Coinbase’s L2 is expected to become the most widely used by 2026. Its massive user base and strong compliance profile make it the biggest onboarding funnel.
2. Optimism
Strong Superchain governance influence and continued development momentum. Their batcher upgrades to rely primarily on blobs cut DA costs by more than half.
3. Arbitrum
Still leading in TVL and DeFi activity. Developer tooling remains strong.
4. zkSync Era
ZK-proof innovation continues. If ZK tech matures as expected, this could be the dark horse.
What This Means for Builders
If you’re building on Ethereum in 2026:
- L2-first is mandatory: Mainnet is for settlement and security, L2s for user activity
- Cost structures change: With more blob space, L2 fees approach near-zero
- UX improvements: Sub-second finality on some L2s becomes viable
- Interoperability matters: Cross-L2 communication becomes a competitive advantage
The Risks
Let’s be real about what could go wrong:
- Upgrade delays: Ethereum’s upgrade schedule often slips
- Validator risks: ePBS introduces new attack surfaces
- L2 fragmentation: Too many L2s could harm composability
- MEV centralization: Even with ePBS, sophisticated actors may still dominate
My 2026 Predictions
- Ethereum mainnet settles $1T+ in daily value across L2s
- Base becomes the #1 L2 by active addresses
- At least one L2 achieves >10,000 TPS sustained
- L2 fees drop to <$0.001 for most transactions
The infrastructure layer is finally catching up to the vision. 2026 is when Ethereum’s rollup-centric roadmap reaches critical mass.
What L2s are you most excited about for 2026?