Hyperliquid Opens Its Prediction Markets to Anyone
HIP-4 goes permissionless: for 500K HYPE, you can now build your own outcome markets on the chain that already ate DeFi's lunch.
By The crypto-money Desk
Until recently, if you wanted to list a prediction market on Hyperliquid, you waited for validators to decide. They picked the questions, set the terms, and hit deploy. That changed when HIP-4 permissionless deployment hit mainnet — and now any team with 500K HYPE can launch its own outcome markets without asking anyone's permission. [1]
The rules are straightforward, if not cheap. That 500K HYPE stake is locked for six months and is slashable if a deployer defines or settles a market poorly. [1] In exchange, deployers get access to a system that currently supports 100 concurrent outcomes and up to 500 new outcomes per day — caps that Hyperliquid has already signalled it intends to raise to 1,000 concurrent outcomes and 5,000 daily deployments. [1]
Deployers can collect up to 50% of the trading fees generated by their markets. [2] The assistance fund converts trading fees to HYPE in a fully automated manner as part of the L1 execution, and HYPE in the assistance fund is then burned, removing the tokens permanently from the circulating and total supply. [2]
What makes this moment significant is less the mechanics than the shift in who holds the pen. Governance by committee tends to move at the speed of consensus. What replaces it here is something closer to a marketplace of market-makers — each one competing to define, fill, and settle questions that traders actually want to trade.
That said, permissionless does not mean unconstrained. Only markets fitting into predetermined outcome templates, approved by validators, can go live. [1] Seven template records are currently approved on mainnet: binaryPrice, priceTouch, scalarPrice, policyRateDecision, policyRateIncrease, policyRateDecrease, and policyRateNoChange. [1] Functionally, that is fewer than seven distinct market types, since the policy-rate templates work together to create multi-outcome questions. [1]
The template constraint means that, for now, allowed markets revolve around finance, spanning price action and policy decisions rather than also including the sports markets that catalyzed HIP-4's growth during the World Cup. [1]
[1] To mark the launch, Outcome unveiled a $1M liquidity program, beginning with $50K in USDC rewards per week across eligible markets, though it remains unavailable to U.S. users. [1]
[1] Skew's opening act was a BTC hourly Up/Down market, though it has since begun rapidly deploying others. [1] Crucially, overlap between deployers is permitted — different builders can offer effectively identical markets alongside one another. [1]
Competition between deployers running parallel markets on the same question is an unusual design choice. In traditional finance, identical products on competing venues fragment liquidity and tend to compress margins until only the largest survive. In a permissionless on-chain context, the dynamics are different — bootstrapping liquidity is the hard part, and letting multiple teams attempt the same question simultaneously is arguably the fastest way to find out who can do it.
Existing validator-deployed markets are not going away. Canonical markets will not be sunset, and their offerings could even expand in the future, though Hyperliquid has said validators should ideally deploy fewer than 10 outcomes or questions per year. [1] Fees on validator-deployed markets have now been switched on, contributing to HYPE buybacks and burns. [1]
Outcome trading carries its own fee structure. Fees are only charged when closing or settling outcome positions, not when opening them. [2] Outcome trading does not support rebates — users who would receive rebates for spot and perp trading pay zero fees on maker orders instead. [2] Markets are denominated in USDC, meaning every dollar flowing into HIP-4 is a dollar parked on Hyperliquid in stablecoin. [1]
The near-term roadmap is readable in the template gaps. Sports are the most visible absence — they were one of HIP-4's biggest use cases during its validator-deployed phase, and football and basketball seasons are approaching. [1] Trade.xyz and Unit Labs have been moving HYPE in 500K-sized chunks in a way that strongly suggests a preparation for HIP-4, though no deployment has been confirmed. [1]
Election markets are the bigger prize. [1] If HIP-4 captures even a modest fraction of that demand — assuming election templates reach mainnet in time — it would represent a step-change in Hyperliquid's prediction market scale. The most obvious catalyst from here is expanding the template set. [1]
What the architecture achieves, structurally, is a separation of concerns. Validators set the rules of the game. Deployers compete to build the best version of it. Traders vote with their capital. Each layer is incentivized to perform its role well, and failure at any layer — a bad settlement, a thin book, a poorly scoped question — has a direct financial consequence for the party responsible.
The removal of the validator bottleneck means Hyperliquid no longer has to decide and deploy virtually every market users might want to trade. [1] The question now is how quickly validators expand the template set and which deployers are best positioned to capitalize when they do. Football, basketball, and the midterms give us three very obvious chances to find out. [1]
Prediction markets have spent years on the regulatory margins, running on specialist platforms with thin liquidity and uncertain settlement. Hyperliquid is not solving the regulatory problem — U.S. users are still locked out of Outcome's rewards program — but it is solving the infrastructure one. [1] A chain that can clear perps at institutional depth can now host outcome markets built by anyone willing to post the stake and define a fair question. The template is open. The calendar is full.