Jupiter (JUP DAO)

Jupiter is the dominant swap aggregator on Solana and the issuer of the JUP governance token. Its governance story is unusually instructive because it runs to both extremes: Jupiter built one of the more thoughtful token-voting systems in the industry – it pays quarterly rewards to holders who actually vote rather than merely hold – and then, in June 2025, it paused that governance entirely, telling its own community that the current structure was not working and that trust had broken down. Few DAOs make the case against their own governance as plainly. This page covers how JUP governance is meant to work, the buyback program that now anchors the token, and the 2025 pause. Every claim links to Jupiter’s own documentation or primary reporting.

Solana’s aggregator and one of its biggest airdrops

Jupiter routes trades across Solana’s liquidity venues to find the best execution, and by volume it is the chain’s leading aggregator. The JUP token launched on 31 January 2024 with a total supply of 10 billion, distributed 40% to airdrops, 40% to the team and investors, 10% to a treasury, and 10% to a foundation (Blockworks). The first airdrop sent roughly 1.35 billion JUP to nearly a million wallets – one of the largest distributions in Solana’s history – and the project has continued the “Jupuary” airdrops in subsequent years (The Block). The scale matters for governance: a token spread across a million wallets is exactly the setting in which low turnout and capital-weighted voting become the central problem, and Jupiter’s design choices are a direct response to it.

Active Staking Rewards: paying holders to actually vote

To vote in the Jupiter DAO you first stake JUP, which locks the token in the governance contract and grants voting power (Jupiter documentation). Staked holders who participate in votes then earn Active Staking Rewards (ASR), distributed each quarter and weighted by how long the stake has been held. Crucially, ASR rewards voting, not mere holding: a staker who never votes forfeits the rewards, which is Jupiter’s answer to the voter-apathy problem that hollows out most token DAOs. Unstaking triggers a mandatory 7-day cooldown that “cannot be skipped or accelerated” – a deliberate guard so nobody can stake just before a distribution, claim, and unstake without any sustained commitment. The rewards pool has run at roughly 50 million JUP per quarter (vote.jup.ag). It is one of the more coherent attempts in the industry to make participation, rather than balance, the thing that governance pays for.

The Litterbox: turning protocol fees into buybacks

At its “Catstanbul” event on 26 January 2025, founder Meow announced that Jupiter would direct 50% of protocol fee revenue to buy JUP back from the open market, holding the repurchased tokens in a long-term reserve nicknamed the “Litterbox”; the other half funds growth and operations (The Block). By late 2025 the Litterbox had accumulated on the order of 130 million JUP, around 4% of circulating supply, and as part of a 2025 “Fresh Start” the DAO put the repurchased stack up for a burn vote (Jupiter). A fee-funded buyback places Jupiter in the same lineage as protocol-owned-treasury designs: value is meant to accrue to the token from real revenue rather than from emissions – the mirror image of the inflationary reward that funds ASR.

When a DAO pauses itself

On 20 June 2025, Jupiter did something few large DAOs have done: it halted all DAO voting until 2026, stating that “the current DAO structure isn’t working as intended” and citing a breakdown in trust between the DAO, holders, and team. Team member Kash Dhanda described the ecosystem as “stuck in a negative feedback loop” (CoinDesk; DL News). No new DAO-funded Work Groups would be created, though ASR continued at its 50-million-JUP-per-quarter rate so that stakers were not penalized for the pause.

Jupiter framed the halt as a reset rather than a retreat – a “Fresh Start” to scale back a governance apparatus that had grown adversarial and rebuild it on simpler terms. It is a candid admission that token-holder governance, even with paid participation and one of Solana’s deepest communities, can become a net drain on a project rather than a source of legitimacy – the failure modes playing out at one of the ecosystem’s most-watched DAOs.

The restart, and what did not come back

Governance resumed in early 2026, but not as it was. Jupiter disbanded every Work Group and narrowed the DAO's remit to proposals that touch tokenomics and treasury directly – revenue allocation, supply control, long-term token strategy. The everyday budgeting the old DAO spent most of its votes on is simply no longer a governance question. That is the reset in one line: fewer things are voted on, and the things that remain are the ones where a token vote is hardest to argue against.

The first real test came in February 2026 with “Going Green”, a proposal to drive net-new JUP emissions to zero for the rest of the year by postponing the Jupuary airdrop, pausing team vesting and buying out the legacy Mercurial vesting. Holders were given the status quo as the explicit alternative. Voting ran 17–21 February and closed on 22 February with roughly 75% for Net Zero and 25% for continuing the airdrop; more than 81% of large-holder voting weight backed the same side. The 2026 Jupuary allocation was cut from a planned 700 million JUP and the withheld tokens returned to Jupiter's community-controlled multisig.

It is worth being precise about what that vote demonstrates, because it cuts both ways. A DAO that had declared its own governance broken came back and used a token vote to cancel a payout to its own token holders – the outcome a purely rent-seeking electorate is supposed to be incapable of reaching. But the whale-weight figure tracking the headline result so closely is the older problem restated: a 75% result that is also an 81% large-holder result tells you less about the electorate than it looks like it does. The narrowing is what changed; the capital-weighted basis of the vote did not.

How Caper approaches this

Jupiter did more than most DAOs to make governance work: it paid holders to vote through ASR, and still concluded that its DAO had become a liability worth pausing. Participation had to be bought with fresh emissions because nothing in the vote itself rewarded it, and a $JUP-weighted vote tracks capital rather than commitment.

In a caper, a member's vote weight comes from the tokens they hold together with the soulbound votes they earn by casting ballots, so participation is part of the weight and needs no separate subsidy. Buying, selling and transferring tokens earn no votes, and nobody can buy, borrow or farm another member's. Holdings still count, but a big bag alone can't capture control. The same weight sets a member's fair share of the treasury, and a member who loses trust can exit any time with it rather than wait for the whole DAO to pause.

References

• Jupiter DAO governance portal – staking, voting, and Active Staking Rewards.
• Jupiter documentation – Staking – the 7-day unstake cooldown and quarterly time-weighted ASR.
• Blockworks and The Block – the 31 Jan 2024 launch, 10 billion supply, and airdrop allocation.
• The Block – the 26 Jan 2025 announcement that 50% of fees fund JUP buybacks (the “Litterbox”).
• CoinDesk and DL News – the 20 June 2025 pause of DAO voting until 2026.