Solana Begins Public Testing of Alpenglow Upgrade Targeting 150-Millisecond Finality

Solana has moved its planned Alpenglow protocol upgrade to the network’s public testnet, where developers will test whether transaction finality can fall from roughly 12.8 seconds to 150 milliseconds. The test is an important migration step, but it is not a mainnet launch.

Alpenglow Reworks Solana’s Consensus Process

Solana currently relies on TowerBFT to reach consensus. Validators record votes on the blockchain and accumulate enough support across 32 slots before a block becomes final, meaning users may wait about 13 seconds before a transaction is considered irreversible.

Alpenglow would replace TowerBFT with Votor, a voting protocol designed to let validators communicate directly and finalize blocks after one or two rounds of voting. The change is intended to shorten the time required for the network to agree on a permanent transaction record without changing how users send funds.

The upgrade has already spent more than four months running on a smaller network created specifically for its development. Moving Alpenglow to Solana’s established public testnet will allow developers to examine whether the wider set of computers and services operating the blockchain can migrate together. Because the testnet uses tokens with no monetary value, teams can identify issues and restart the network without putting user funds at risk.

Testnet Migration Has Important Limits

Validators participating in the initial test must run Agave 4.3, the primary Solana validator software maintained by Anza. Anza recommended the release to mainnet validators on September 21, after first distributing it to operators representing 10% and then 25% of the SOL securing the network.

The first Alpenglow test does not yet support Firedancer or Frankendancer, alternative validator clients developed by Jump Crypto. Those projects are intended to reduce Solana’s dependence on a single software implementation, limiting the risk that a fault in Agave could affect the entire network at once.

As a result, the initial migration will run entirely through Agave. That makes the test useful for evaluating the upgrade’s core behavior, while leaving broader client diversity and compatibility for later stages.

• Current finality target: approximately 12.8 seconds
• Alpenglow target: approximately 150 milliseconds
• Initial required validator software: Agave 4.3
• Unsupported clients in the first test: Firedancer and Frankendancer

What This Means

If testing succeeds, Alpenglow could materially improve Solana’s usefulness for applications that depend on rapid settlement, including payments, trading and other real-time on-chain services. Faster finality may also strengthen the network’s competitive position against blockchains promoting near-instant confirmation.

However, the public testnet is still a validation phase. September 28 appears in Anza’s schedule as a tentative date for activating features included in Agave 4.3 on mainnet, but it is not a confirmed Alpenglow launch date. The upgrade’s timeline remains dependent on testnet results, migration readiness and support from additional validator clients.