Hyperliquid Is Opening Prediction Markets to Anyone. ‘Anyone’ Has a Price

Hyperliquid wants anyone to be able to launch a prediction market.

That sentence needs an asterisk the size of the deposit.

The planned permissionless version of HIP-4 requires each deployer to stake 500,000 HYPE. At roughly $60 per token, that is about $30 million locked behind the right to create markets.

The stake can be slashed.

A market that is poorly defined, settled incorrectly or left unresolved for more than a week can cost the deployer part or all of that capital if validators vote to take it.

Hyperliquid is opening the door. It is also making sure almost nobody walks through it casually.

Hyperliquid’s announcement says permissionless deployment will arrive in a future network upgrade, first on testnet and then on mainnet after builders and validators test the new controls. The change expands HIP-4, the fully collateralized outcome-trading system launched in May.

Today, validators control which outcome markets appear. The new model pushes most of that work to outside deployers while keeping validators in charge of the templates, sanctions and a small set of canonical markets.

The result is neither a free-for-all nor a traditional exchange listing committee. It is a franchise system enforced by code and collateral.

Hyperliquid says validator-created canonical markets should become rare, ideally fewer than 10 outcomes or questions each year. The wider catalog would come from deployers using approved structures.

The specifications remain preliminary and can change after testnet feedback, so the capital rules and launch sequence are a design commitment rather than a finished mainnet product.

That staged rollout gives validators and builders a chance to test settlement disputes before unrestricted mainnet deployment begins.

LATEST: ⚡ Hyperliquid plans to open HIP-4 outcome markets to permissionless deployment, with deployers required to stake 500,000 HYPE.

— CoinMarketCap (@CoinMarketCap) July 20, 2026

Templates come first.

A prediction market sounds simple: ask a question, sell contracts tied to possible answers, and pay the winning side after the event.

The dangerous part is the sentence between the question and the payout.

What counts as an official result? Which source controls settlement?

What happens if an election is delayed, a game is canceled or two government agencies publish conflicting numbers?

Those are not editorial details. They determine who gets the money.

The Block reports that Hyperliquid validators will approve standardized outcome templates whose rules are stored and enforced onchain, giving every market in a template family the same basic contract structure. Deployers can then use those approved templates to create individual markets and define their specific settlement terms.

Validators retain the power to punish bad execution. A deployer who writes an ambiguous market or settles against the stated criteria puts the 500,000-HYPE bond at risk.

The stake remains locked for six months. Outstanding markets have to be settled before a deployer can unstake.

Each deployer starts with capacity for 100 outcomes, or 200 outcome tokens. Multi-answer questions use more of that allocation, while completed markets release capacity for another use.

Hyperliquid plans a later auction for teams that want more slots, adding another price signal to the market-creation layer.

Those limits give the network a controlled way to expand without letting one deployer crowd the order book with endless questions.

That time lock matters more than the headline dollar value. HYPE can move sharply during six months, and the deployer carries that market exposure while also operating the prediction-market business.

The bond is therefore capital, collateral and reputation at the same time.

🔓 Hyperliquid wants to blow open its prediction markets. Stake 500,000 HYPE (~$30M) and anyone can deploy one, no validator gatekeeping required.

— Laura Shin (@laurashin) July 20, 2026

The reward is a piece of the market.

Each deployer is expected to receive capacity for 100 outcomes, represented by 200 outcome tokens. Multi-answer questions consume more of that allocation, while settled outcomes free capacity for reuse.

Hyperliquid also plans an auction mechanism for deployers that want a larger allocation.

The business model sits in the fee switch. Deployers may set fees of up to 50% on their markets, giving them a direct reason to create questions people actually want to trade.

That is a powerful incentive and a potential distortion.

A deployer earns more when volume rises. The most lucrative questions may be the ones with the strongest emotion, biggest audience or sharpest information divide—not necessarily the ones with the greatest public value.

The slashable stake is meant to keep that incentive from turning into a flood of sloppy or manipulative contracts.

It also makes “permissionless” a technical description, not an accessibility promise.

Anyone can deploy without asking a validator to approve each individual question. Very few teams can lock roughly $30 million of HYPE, build settlement operations and accept the risk of a validator vote against them.

That is likely intentional.

Prediction markets have a spam problem before they have a scale problem. Hundreds of near-duplicate questions fragment liquidity.

Tiny markets produce weak prices. Ambiguous contracts create settlement fights that can damage trust across the whole venue.

A giant bond filters the applicant pool down to teams with capital, infrastructure and something substantial to lose.

Decrypt notes that HIP-4 generated roughly $176 million in World Cup-related volume while the wider prediction-market sector handled tens of billions during the tournament across established venues. Hyperliquid has demand, but it remains far behind the category’s largest competitors.

Permissionless deployment is the bet that outside builders can close that gap faster than a validator committee selecting markets one at a time.

The World Cup numbers also show the size of the opening. Hyperliquid found product demand, yet Kalshi and Polymarket captured the overwhelming share of a tournament-driven boom.

Outside deployers can specialize by sport, region or event type in ways a central team may never prioritize. Their fee revenue rises only if traders show up.

The comparison also sets a hard benchmark: broader listings have to create durable liquidity, not a long menu of empty contracts.

Hyperliquid expects validator-created canonical markets to become rare—ideally fewer than 10 outcomes or questions per year. Most expansion would come from deployers working from approved templates.

That division of labor is the real upgrade.

Validators write the grammar. Deployers write the questions.

Traders decide which ones deserve liquidity.

The model could produce a much broader market catalog without surrendering settlement standards. It could also concentrate market creation among a handful of wealthy operators whose incentives look a lot like those of an exchange.

The testnet will show whether the structure can balance those forces before real capital depends on it.

Hyperliquid is making market creation open in the narrowest, most consequential sense: no one has to ask permission.

They just have to put $30 million on the table and be right.

Join the conversation!

We have no tolerance for comments containing violence, racism, profanity, vulgarity, doxing, or discourteous behavior. If a comment is spam, instead of replying to it please click the icon below and to the right of that comment. Thank you for partnering with us to maintain fruitful conversation.