Solana has presented Quantumglow, a proposed post-quantum adaptation of its newly activated Alpenglow consensus, as a way to protect validator agreement against future quantum attacks without sacrificing the network’s fast finality. The proposal was detailed in a Solana Developers video explainer on Oct. 22, featuring Jacob Creech and Quentin Kniep, and shared by Solana’s main account, which said the network “has been early in quantum readiness and remains ahead in its preparation.”
Quantumglow is not yet a scheduled upgrade. Neither the video nor the accompanying material provided an activation date or a Quantumglow-specific Solana Improvement Document. The underlying research was published earlier by infrastructure firm Anza on July 30, authored by Quentin Kniep, Kobi Sliwinski and Roger Wattenhofer, and points to Section 5 of the updated Alpenglow whitepaper.
The main technical challenge is that post-quantum signatures are larger, while the compact signature aggregation used to prove validator agreement has no ready replacement under the researchers’ requirements. To address this, Anza proposes Ax, short for Alpenglow XMSS, a hash-based signature scheme tailored to the protocol. Validators would exchange and cache one-time public-key material ahead of use, reducing the data that must accompany later signatures. For block authentication, the design uses a signed commitment covering the block payload, supported by authenticated communication channels, rather than attaching a large post-quantum signature to every piece of block data. A related distributed-quorum-signature paper, submitted July 20, describes moving consensus certificates from network messages to locally recorded events once a node gathers enough evidence. The researchers claim each message remains constant in size regardless of node count, but note total communication grows quadratically, with no measured mainnet result yet.
Separately, Solana Foundation’s upgrade page confirms Alpenglow went live on mainnet during epoch 1064 on Oct. 21 at slot 459,653,000, supported by votes from 446 of 654 validators. The Foundation lists a finality target of about 150 milliseconds, a major reduction from the 12.8 seconds under TowerBFT. The deployed portion is the Votor voting system under SIMD-0326; Rotor, a data-distribution component, is reserved for a later release. The Foundation says Alpenglow leaves transaction execution, programs and transaction fee mechanics unchanged, and ordinary users do not need to change how they sign, send or approve transactions. Analytics providers, however, must reset metrics that previously counted validator votes alongside user transactions.
Quantum security work extends beyond consensus. Anza’s April 27 assessment by Max Resnick and Sam Kim identified quantum vulnerabilities in accounts, block propagation, consensus and user-defined programs, warning that an attacker able to break Ed25519 could forge account signatures or support conflicting confirmations. In late April, Anza and Firedancer also tested the Falcon signature scheme as a possible future option. On Bitcoin, BIP-360—a draft by Hunter Beast, Ethan Heilman and Isabel Foxen Duke—proposes Pay-to-Merkle-Root outputs through a soft fork to remove Taproot’s key-path spending and protect long-exposed public keys, though its authors note that protecting transactions waiting for confirmation may require post-quantum signatures in a separate proposal.
