Ethereum’s Fusaka upgrade went live on 3 December 2025. It was the second major upgrade of 2025 after Pectra and changed how developers, infra providers, rollups, and anyone pushing transactions through an Ethereum RPC or blockchain data API should size gas and blobs.

Fusaka brings three core shifts:

More L1 throughput: block gas limit rises to ~150M gas, pushing L1 capacity into the 40-60 TPS range

A hard per-transaction gas cap: EIP-7825 limits single tx gas to 2²⁴ = 16,777,216 gas

Smarter data availability economics: PeerDAS + EIP-7918 scale blob capacity while introducing a blob fee floor tied to L1 gas costs

Below is a clean breakdown of what changed and what you should still verify in your stack.

EIP What It Does Impact Notes
EIP-7825 Caps per-transaction gas at 2²⁴ (~16.78M) Execution safety Prevents mega txs, enables parallel execution
EIP-7918 Introduces blob base fee floor tied to L1 gas DA pricing stability Ends near-zero blob fees, aligns L2 costs with L1
EIP-7917 Deterministic proposer lookahead Based rollups Unlocks preconfirmations, validator sequencing
EIP-7939 Adds CLZ (Count Leading Zeros) opcode EVM efficiency Improves bitwise ops, packed data, zk patterns
EIP-7951 Native secp256r1 (P-256) signature verification Better UX Enables WebAuthn, passkeys, hardware wallets
EIP-7907 Allows larger contract code with scaled gas cost Bigger contracts Supports complex dApps without proxy tricks
PeerDAS Peer-based blob data sampling More blobs Raises DA throughput, reduces node load

EIP-7825: The 16.78M Gas Cap

Before Fusaka, a single transaction could consume an entire block. That is incompatible with parallel execution and creates DoS risk.

After Fusaka:
Every transaction is capped at 2²⁴ gas (≈16.78M), independent of the 150M gas block limit.

Who is affected:

-Heavy deployments

-Large batch operations

-Routers/aggregators that chain many calls in one tx

What to do:

-Simulate deployments and batch jobs on Hoodi or Sepolia (Holesky is deprecated)

-Split large operations across multiple transactions

-Update any builder/wallet logic that allows tx gas above 16,777,216

eth_call is unaffected, only confirmed txs enforce the cap.

Block Gas Limit → 150M

Fusaka increases the block gas limit roughly 3x, enabling 40-60 TPS depending on workload.

Impact:

-More room for L2 settlement + normal L1 activity

-More data to index per block

-Higher CPU, bandwidth, and disk requirements for node operators

If you run your own Ethereum nodes or rely on a provider, confirm their Fusaka-ready client versions (Geth, Erigon, Reth, Nethermind, Besu).

PeerDAS + EIP-7918: Blob Capacity Scaling and Fee Floors

Fusaka reshapes Ethereum’s data availability layer, which powers rollups.

PeerDAS

Instead of every node downloading full blob data, validators sample pieces from peers, drastically reducing per-node load. This unlocks higher blob counts (10→14 per block shortly after activation).

EIP-7918: Blob Base Fee Floor

Blob fees can no longer sit near zero while L1 gas is high. They now follow a minimum fee linked to L1 execution cost.

A backtest shows the new rules would have generated:‍

-≈24,641 additional ETH in blob revenue

-Blob prices higher on ≈93% of days since Dencun

This redirects more value to ETH stakers and makes L2 DA costs more predictable, though slightly higher.

Action for rollups and DA-heavy apps:

-Update blob cost models

-Adjust batch sizes

-Monitor blob usage around activation

UX-Focused EIPs (Verkle Was Not in Fusaka)

Several smaller but impactful improvements landed in Fusaka:

EIP-7951 → native P-256 verification for WebAuthn and hardware-secure logins

EIP-7939 → new CLZ opcode for efficient bit operations

Verkle trees were not activated in Fusaka. Compact Verkle state proofs remain a later Ethereum upgrade. If you build wallets, zk systems, AA flows, or mobile clients, EIP-7951 and EIP-7939 are the Fusaka items that matter now.

EIP-7917: Deterministic Proposer Lookahead & Based Rollups

Validators can now see in advance who will propose future blocks.
This unlocks:

-Preconfirmations

-Based rollups, where Ethereum validators sequence L2 transactions

-New MEV/tip flows that accrue to ETH stakers instead of centralized L2 sequencers

If you operate or design a rollup, expect roadmap conversations around integrating preconfirmations.

Test Fusaka on Live Networks

Run gas-cap, blob, and indexing checks against Fusaka mainnet behavior. Hoodi and Sepolia are the current Ethereum testnets. Get your API key and hit Ethereum endpoints through Tatum’s gateway.

Get API Key Here

Builder Checklist After Fusaka

Smart Contracts & Scripts

-Ensure no tx exceeds 16,777,216 gas

-Replay heavy operations on Hoodi or Sepolia

-Break down deployments or migrations into smaller calls

SDKs, Wallets, Transaction Builders

-Add client-side validation for the new gas cap

-Update gas estimation logic

-Handle failures when a transaction would exceed the gas cap

Nodes, RPC, and Data APIs

-Benchmark block ingestion assuming 150M gas blocks

-Update to Fusaka-compatible client releases

-Monitor blob usage and fee fluctuations

Rollups & DA Users

-Recalculate DA costs with the blob fee floor

-Test PeerDAS-driven blob throughput increases

-Evaluate based rollup strategy and preconfirmation UX

If you rely on Tatum RPC, Data API, or Notifications, point staging at Hoodi or Sepolia so you can measure gas patterns, blob usage, and indexing performance on current testnets.

Final Thoughts

Fusaka realigns Ethereum around three priorities:
scale L1, scale DA, improve UX.

It increases capacity, stabilizes blob pricing, removes harmful “mega-transactions”, and sets the foundation for based rollups. For developers, the main task is updating assumptions, especially around gas ceilings and blob behavior.

If you do that, your contracts, rollups, and infra stay aligned with Fusaka mainnet rules.