Starknet v0.14.3 Mainnet Launch Adds Dynamic L2 Gas

6 mins mins

Starknet is preparing to launch v0.14.3 on mainnet on 22 Jun 2026, introducing a dynamic L2 gas adjustment mechanism that automatically tunes the minimum base fee between blocks. The upgrade also cuts target L2 gas per block by 30% and shortens block times, marking one of the more significant fee-model changes for the Ethereum Layer-2 network this year.

Starknet v0.14.3 Mainnet Launch Adds Dynamic L2 Gas

What Starknet v0.14.3 Will Introduce on Mainnet This Month

The v0.14.3 pre-release notes published on the Starknet community forum outline a two-phase rollout. Testnet activation is scheduled for 9 Jun 2026, with mainnet following on 22 Jun 2026.

Mainnet rollout date

22 Jun 2026

Official pre-release notes place Starknet v0.14.3 mainnet deployment on 22 Jun 2026.

The centerpiece of the release is SNIP-35, a new algorithm that automatically adjusts the minimum L2 gas base fee between blocks. Alongside SNIP-35, the upgrade reduces target L2 gas per block by 30% while keeping the maximum block size unchanged.

Target L2gas reduction

30%

The release tightens the target L2gas per block by 30%, a core part of the new dynamic fee adjustment design.

Block times are also being shortened in this release. The combination of faster blocks, a lower gas target, and dynamic fee adjustment represents a coordinated redesign of how Starknet manages throughput and pricing pressure.

How Dynamic L2 Gas Adjustment Changes Transaction Pricing

Starknet’s current fee model operates similarly to Ethereum’s EIP-1559 mechanism, targeting roughly 80% block capacity with a minimum price floor of 3 gFri. Under this system, fees adjust based on how full blocks are relative to the target, but the minimum base fee has remained static.

SNIP-35 changes this by making the minimum L2 gas base fee itself variable. Instead of a fixed floor, the algorithm evaluates network conditions between blocks and raises or lowers the minimum fee automatically. This means that during sustained high demand, the floor rises to discourage spam, while during quiet periods it can drop to keep transactions cheap.

By simultaneously lowering the target L2 gas per block by 30%, the upgrade makes blocks reach the “full” threshold sooner. This triggers the EIP-1559-style price adjustment more aggressively, giving the network a tighter feedback loop between demand and cost. The maximum block size stays the same, preserving burst capacity for sudden traffic spikes.

The net effect is a fee market that responds more quickly and granularly to actual usage patterns rather than relying on a static minimum that may be too high during low activity or too low during congestion.

Why the v0.14.3 Upgrade Matters for Users and Developers

For end users, the dynamic fee model should make transaction costs more predictable in relative terms. Rather than encountering sudden fee jumps when blocks cross the target threshold, fees will adjust incrementally between blocks. During periods of low network activity, users could see lower minimum fees than the current fixed floor allows.

Developers and infrastructure operators face a more immediate action item. The official Starknet version page confirms that RPC 0.8 is no longer supported in v0.14.3. Wallets, RPC providers, explorers, and indexers need to migrate to RPC 0.9 at minimum, with RPC 0.10.1 or 0.10.2 recommended.

This deprecation means that any dApp or service still running on RPC 0.8 will break after the mainnet rollout. Teams building on Starknet have roughly two weeks from testnet launch on 9 Jun to validate compatibility before mainnet activation. In a market environment where the Bitcoin premium index has been negative for 19 days, infrastructure stability is especially important for retaining developer confidence.

The shorter block times introduced alongside the fee changes also affect applications that rely on block-based timing. Smart contracts using block numbers as time proxies will need to account for the faster cadence.

What This Means for Starknet in the Ethereum Layer-2 Landscape

Starknet currently holds approximately $204 million in total value locked, positioning it as a mid-tier Layer-2 by TVL. The STRK token trades at roughly $0.031 with a market cap near $198 million, reflecting broader market weakness as the Fear & Greed Index sits at 12, deep in “Extreme Fear” territory.

Fee-market efficiency is becoming a competitive differentiator among Ethereum Layer-2 networks. As rival chains pursue their own infrastructure overhauls, including Solana’s Alpenglow consensus upgrade currently in testing, Starknet’s move toward dynamic gas pricing addresses one of the persistent criticisms of Layer-2 fee structures: that static minimums create unnecessary costs during low-demand periods.

The 30% target gas reduction is particularly relevant for adoption. Lower effective costs per transaction make Starknet more competitive for high-frequency use cases like gaming and micropayments, segments where even small fee differences influence which chain developers choose. In an environment where broader crypto markets have shown no evidence of mass selling despite macro uncertainty, protocol-level improvements like this represent the kind of fundamental progress that can attract builders regardless of token price action.

The upgrade does carry adjustment risk. A more aggressive fee curve could temporarily disrupt MEV strategies and gas-estimation tooling that was calibrated to the old parameters. The testnet window between 9 Jun and 22 Jun gives the ecosystem time to surface these issues before they affect mainnet users.

FAQ About Starknet v0.14.3 and Dynamic L2 Gas Adjustment

When is Starknet v0.14.3 expected on mainnet?

The official pre-release notes schedule mainnet activation for 22 Jun 2026, with testnet going live on 9 Jun 2026.

Will users see lower transaction fees?

Potentially. The dynamic adjustment allows the minimum base fee to drop during low-demand periods, which is not possible under the current fixed-floor model. However, during high demand, fees could also rise more aggressively than before due to the lower gas target per block.

Do developers or users need to take action?

Developers and infrastructure operators running RPC 0.8 must upgrade to at least RPC 0.9 before mainnet launch, with RPC 0.10.1 or 0.10.2 recommended. Regular users transacting through wallets and dApps should not need to take manual action, but may notice changes in gas estimation displays.

What is SNIP-35?

SNIP-35 is the specific proposal that introduces the automatic algorithm for adjusting the minimum L2 gas base fee between blocks. It is the core technical component of the v0.14.3 release.

Does the maximum block size change?

No. The maximum block size remains unchanged. Only the target L2 gas per block is reduced by 30%, which affects how quickly the fee-adjustment mechanism activates, not the absolute limit on block capacity.

Disclaimer: This article is for informational purposes only and does not constitute financial or investment advice. Cryptocurrency and digital asset markets carry significant risk. Always do your own research before making decisions.