Starknet vote on leaving Ethereum: who is actually voting

The Starknet vote is a non-binding Snapshot poll that asks STRK holders whether Starknet should research leaving Ethereum to become an independent post-quantum Layer 1. It opened on October 15, 2026 and closes on October 22 at 09:32 UTC. Yes leads with 99.96% of the voting power. When I pulled every vote from Snapshot, 94.1% of that power turned out to be STRK the Starknet Foundation had delegated to 41 addresses.

Holders voting with their own STRK cast 22.0 million in voting power, about 1.4% of the STRK that is staked today. They also voted Yes, by 99.3%, so the result would be the same without the Foundation's delegates. What the headline number hides is how few people decided it.

Starknet vote: what the poll actually asks

The poll has three choices: "Yes, this should be researched", "No, this shouldn't be researched" and "Neutral, I need more information". The proposal text says a Yes "indicates support for further research" and "does not constitute approval of a transition to an L1". Any real move would need its own governance process, and the poll has no quorum.

Two things are fixed whatever the result. In the forum post that goes with the poll, the Starknet Foundation writes that STRK stays the token, with no new token, and that the bridges from Ethereum to Starknet stay open. If Yes wins, a research effort starts and leads to more detailed proposals and further polls. If No wins, Starknet keeps its post-quantum work as a Layer 2. A Neutral result keeps the forum thread open until the missing answers are in.

The headline "Starknet leaves Ethereum" is ahead of the text. What STRK holders are approving this week is a research budget and a direction. The forum post itself says that no timeline for an L1 exists yet.

Why does Starknet want to leave Ethereum?

Starknet is a Layer 2: it executes transactions itself, then proves them with STARK proofs and settles them on Ethereum, where it also publishes its state data. The STARK proofs are hash-based, which is why StarkWare calls the core protocol safe from quantum computers. The weak points sit in what Starknet borrows from Ethereum, and the forum post names two of them: the bridge and data availability.

A Layer 2 can only be as quantum-safe as its base layer. That is the argument StarkWare's CEO Eli Ben-Sasson made on X on October 8. The same day the Starknet account posted that becoming an L1 would make Starknet the first fully quantum-resistant network, "with 2027 as our target". On October 10 StarkWare called an independent L1 "our clearest path forward".

Ethereum has its own date, and it is later. The Ethereum Foundation's Protocol cluster wrote on September 7 that it aims to make Ethereum quantum-resistant across execution, consensus and data by December 2029, and to treat that deadline as fixed until a reassessment in January 2027. Starknet's case for leaving is the two years between its 2027 target and Ethereum's 2029 one.

The 2027 date belongs to quantum resistance. Several outlets reported 2027 as the date Starknet moves to its own chain, and the posts on X give 2027 as the target for full quantum resistance. The governance post, the only official text behind the vote, gives no date for an L1. If you read "Starknet L1 in 2027" in a headline, it is a paraphrase.

Who cast the votes in the Starknet vote: I counted every one

On October 19 at 08:59 UTC I pulled all 2,593 votes through Snapshot's public API, with each vote's power split by the strategy that produced it. Starknet's space counts four sources of voting power: vSTRK on Starknet, STRK on Ethereum, staked STRK, and a fixed list of addresses labelled "STRK (SNF delegated)". That list is the voting power the Starknet Foundation lends to its governance delegates.

Source of voting power	Voting power	Share
STRK (SNF delegated), 41 addresses	353.70M	94.1%
Staked STRK	21.29M	5.7%
vSTRK on Starknet	0.70M	0.2%
STRK on Ethereum	0.02M	0.0%

The Foundation's list holds 931.5 million in delegated voting power across 109 addresses, in three fixed sizes: 18 addresses with 16,666,667, 34 with 9,259,259 and 57 with 5,555,556. That is why the top of the vote looks so neat: 13 of the 20 biggest voters carry exactly 9,259,259. By October 19, 41 of the 109 had voted, casting 38% of the delegated power.

The other 2,552 voters used their own STRK, almost all of it staked. The median vote carried 107 STRK of voting power. Fifteen addresses voted No, with 158,688 between them, and one voted Neutral.

Holders took part more than they did a year ago. The Foundation published the numbers for its Q3 2025 poll with and without its delegates, which makes a fair comparison possible.

	Q3 2025 poll (average per question)	This vote, Oct 19
Voters	1,202	2,593
Voting power without Foundation delegates	3.5M	22.0M
Delegates' share of voting power	99.0%	94.1%

Six times more own STRK voted this time, and it still came to 1.4% of the 1.54 billion STRK that Staking Rewards shows as staked. The decision runs through the Starknet Foundation's delegates, and for now the holders mostly watch.

What the Starknet L1 plan means for STRK token holders

Your STRK does not change. The Foundation's post says STRK stays the token and no new token is coming, so there is nothing to swap or claim. Treat any site offering a "new Starknet L1 token" as a scam.

You cannot add voting power to this poll now. Snapshot counts balances at Ethereum block 26,196,434, fixed when the poll was created. STRK you stake or wrap into vSTRK today counts in the next poll, not in this one.

The Starknet bridge stays open. Assets bridged from Ethereum sit in contracts on Ethereum either way. If you hold funds on Starknet, the practical step against quantum risk is the account itself. Starknet's account abstraction lets a user move to a post-quantum signature scheme without a protocol upgrade, and the first such accounts are experimental today. A wallet you control is the starting point, and this guide covers the basics.

The STRK token price already went up and came back. By CoinGecko's daily prices, STRK was at $0.049 on October 8, the day StarkWare floated the idea. It peaked at $0.123 on October 12 and was back at $0.059 on October 19. A scheduled unlock of 127 million STRK on October 15, listed on Tokenomist, fell inside that window.

Check the ticker before you search. "STRK stock" and "STRK dividend" in Google are about Strategy's STRK preferred shares on Nasdaq, a different asset with the same four letters.

Starknet staking and the security question

An independent chain needs its own validators to order, confirm and finalize blocks, a job Ethereum does for Starknet today. Staking Rewards shows 137 active Starknet validators and 1.54 billion STRK staked. In its own analysis, PostQuantum valued that stake at about $117 million on October 17 and compared it with roughly $103 billion of staked ETH.

Starknet's staking is in its second phase, where validators attest to blocks. StarkWare targets late 2026 for Staking v3, in which validators start validating blocks. The design for validators signing blocks, SNIP 33, is still a draft.

That is the honest limit of this vote. It measures whether people want the research, and the research is where the security answers would come from. PostQuantum's piece goes deeper into the cryptography than I can, and I would read it before forming a view on the move itself.

What happens after October 22

If Yes wins, which on these numbers it will, the Foundation starts a research effort and comes back with detailed proposals and further polls. The next fixed date on Ethereum's side is the January 2027 reassessment of its December 2029 target. A future vote on an actual move is the one where delegated power and holders' own STRK will matter more, and the split above is the number I will check first.

If you farm or hold tokens from airdrops, governance like this is part of what those tokens are for, and I wrote about how airdrop farming looks from the inside.

Sources

• Snapshot, Should Starknet become an independent post-quantum L1?, proposal text, strategies and all 2,593 votes, read through hub.snapshot.org on October 19, 2026, 08:59 UTC.
• Starknet Community Forum, Should Starknet become an independent post-quantum L1?, October 15, 2026.
• Starknet Community Forum, Q3 Poll Results, December 1, 2025.
• Ethereum Foundation, Protocol priorities, September 7, 2026.
• X posts by Eli Ben-Sasson, Starknet and StarkWare, October 8 to 10, 2026.
• PostQuantum, Starknet Weighs a Post-Quantum L1 Exit From Ethereum, October 2026.
• Staking Rewards, Starknet staking data, read October 19, 2026. STRK daily prices from CoinGecko, October 6 to 19, 2026.