Solana has moved closer to one of its biggest network upgrades. Its planned Alpenglow upgrade has now reached a second public test network. The upgrade aims to cut transaction finality from about 12.8 seconds to around 150 milliseconds.
This is a very large change for the blockchain. A wait of 12.8 seconds may not seem long in normal life, but it matters a lot in crypto. Exchanges, payment apps, trading platforms and blockchain bridges often need to know that a transaction is final before they take the next step.
With Alpenglow, that wait could fall to a small part of a second. The goal is to make Solana transactions reach finality almost at once.
The upgrade is not live on Solana’s mainnet yet. It is still in the test phase. This means developers and validator operators can check how the new system works before any change reaches the blockchain that holds real user funds.
What Is Alpenglow?
Alpenglow is a major change to Solana’s consensus system. Consensus is the process that lets validators agree on the state of the blockchain and decide when a transaction can no longer be reversed.
Solana now relies on a system called TowerBFT. Under this system, validators vote through vote transactions that become part of blocks. A block then needs enough votes across a long confirmation process before it reaches finality. This process takes about 12.8 seconds.
Alpenglow will replace TowerBFT with a new system called Votor.
The main idea is simple. Instead of putting validator votes into blocks as normal transactions, validators can send their votes directly to one another. These votes can then form a certificate that proves enough stake has agreed on a block.
This change removes much of the delay from the old process.
From 12.8 Seconds to 150 Milliseconds
The headline figure is the move from 12.8 seconds to about 150 milliseconds.
One second has 1,000 milliseconds. So 150 milliseconds is only 0.15 seconds. If Alpenglow reaches its target, finality could arrive in less time than most people need to notice a screen response.
The Solana Foundation itself lists the expected change as about 12.8 seconds to 150 milliseconds. It describes Alpenglow as a faster path from a transaction to certainty.
There is also a faster path inside the new system. With enough validator stake, a block can reach finality after one round. A second round acts as a fallback when the first round does not have enough support.
The important point is that 150 milliseconds is a target, not a promise that every transaction will always reach finality in exactly that time. Actual results can vary with network conditions and the amount of validator stake that takes part.
Why Finality Matters
Finality is different from simply seeing a transaction on the blockchain.
When a transaction has finality, the network has reached a strong agreement that the transaction will not be reversed under its consensus rules. This matters for businesses that need a clear point of certainty.
For example, an exchange may wait for finality before it credits a customer’s deposit. A bridge may wait before it releases funds on another blockchain. A merchant may want finality before it treats a crypto payment as complete.
A shorter finality period could make these services feel much faster.
Today, Solana already offers fast transaction execution. Alpenglow does not try to change that basic part of the network. Instead, it changes the way validators reach final agreement.
That difference is important. The upgrade does not require normal users to learn a new way to send SOL or interact with Solana apps.
The Second Public Test Network
Alpenglow has now reached both Solana’s public testnet and devnet. The testnet transition took place on September 24, 2026, while devnet completed its transition on September 25, 2026.
This stage gives developers and validator operators a much wider place to test the upgrade.
A public test network does not hold real user funds in the same way as mainnet. This lets teams find software problems, test validator behavior and check how the new consensus system works at a larger scale.
Before this stage, Alpenglow had spent more than four months on a smaller community test cluster. That earlier work gave developers a place to test the basic migration before the upgrade moved to Solana’s established public networks.
The move to public testnet is therefore an important step, but it is not the same as a mainnet launch.
What Changes for Solana Users?
For most users, very little will change at the surface level.
Wallets can keep the same basic transaction process. Solana programs and the Solana Virtual Machine also remain in place. Transaction formats and fees do not change as part of the Votor phase.
The main change takes place below the user level. Validators will use a different way to reach agreement and create finality.
This means a user may not notice a new button or feature in a wallet. Instead, the benefit should appear as a much shorter wait for final settlement.
Developers whose apps only send transactions or read account data may also need little or no change. Infrastructure that reads block, vote or finality data may need more work.
The Solana Foundation says developers should also prepare for changes to block data, while normal transaction execution remains the same.
Validators Have a Bigger Role
Validators face a more serious change.
The first Alpenglow phase, called Votor, requires validator operators to use Agave 4.3, the main Solana validator software maintained by Anza.
The rollout has taken place in stages. Anza first asked operators with about 10% of staked SOL to adopt the new software, then expanded that recommendation to operators with about 25% of stake. A broader recommendation followed on September 21.
Another point matters here. Firedancer and Frankendancer, alternative Solana validator clients from Jump Crypto, do not yet support the Alpenglow test. That means the first public test phase relies on Agave support.
Client diversity remains an important part of Solana’s wider infrastructure plans, so support from other clients will matter as the upgrade moves ahead.
No Confirmed Mainnet Launch Date Yet
Alpenglow has moved much closer to mainnet, but there is still no confirmed public date for the full upgrade.
Some reports have cited September 28, 2026 as a tentative date for an Agave 4.3 feature activation on mainnet. That should not be treated as a confirmed Alpenglow launch date. The two events are not the same.
The upgrade still needs more tests before the final switch.
This is especially important because the 150-millisecond figure has not yet been proven under full mainnet conditions. Public testnet results can provide useful evidence, but mainnet has real economic activity, real assets and a much larger set of services that depend on the network.
What Comes After Votor?
Alpenglow is planned as more than one change.
The first phase is Votor, which replaces the old validator vote process and aims for about 150-millisecond finality.
A later phase called Rotor is planned as a separate improvement. Rotor is designed to replace Solana’s current Turbine block propagation system with a different network design that could reduce latency further. The Solana Foundation says this phase will come after Votor and has not yet received a fixed schedule.
So the 150-millisecond target is not the end of Solana’s performance work. It is the first major part of a wider plan to reduce delay across the network.
Why This Matters for Solana
Alpenglow could change how fast Solana feels for users and businesses.
The network already has a reputation for fast transaction execution. A finality time of about 150 milliseconds would add a much faster point of certainty to that experience.
For exchanges, payment services, bridges and other apps, the change could reduce the time between a transaction and the next action. For developers, it could also make new types of applications easier to build.
Still, the most important test is ahead. The upgrade must prove that its speed target can work with strong reliability across a large public network.
For now, Alpenglow has passed another major stage. It is active on Solana’s public testnet and devnet, and the target remains clear: move finality from roughly 12.8 seconds to about 150 milliseconds.
If that target holds through the remaining tests and later reaches mainnet, it could mark one of the most important changes to Solana’s core network since its launch.
ALSO READ: Bitcoin Holds $84K as BTC Dominance Falls Below 60%
