Polymarket and Hyperliquid walked onto the TOKEN2049 stage this week as two of the few protocols in the industry that generate material user-paid fees. Combined, they produced approximately $157 million in 30-day gross fees as of early October 2026 — Hyperliquid at $90.6 million and Polymarket at...
Polymarket and Hyperliquid walked onto the TOKEN2049 stage this week as two of the few protocols in the industry that generate material user-paid fees. Combined, they produced approximately $157 million in 30-day gross fees as of early October 2026 — Hyperliquid at $90.6 million and Polymarket at $66.4 million, according to DefiLlama. In a sector where most protocols subsidize usage through token inflation and venture capital, these two stand out because their revenue comes overwhelmingly from traders paying to trade.
That distinction matters. As covered in Parts 1 through 4 of this series, the majority of TOKEN2049 announcements fell into two buckets: shipped products from TradFi firms generating application-layer fees, and narrative from crypto-native projects with no near-term revenue path. Hyperliquid and Polymarket occupy a third category — crypto-native protocols with real, measurable fee revenue at scale. The question is whether that revenue is durable, or whether regulatory risk, fee-sharing dilution, and volume concentration undermine the model.
This is Part 5 of webthreepedia's six-part live series from TOKEN2049 Week.
Hyperliquid founder Jeff Yan appeared across multiple TOKEN2049 sessions, including "Hyperliquid: Rebuilding Finance from First Principles" on Day 1 and the "Building the Infrastructure for All Finance" panel on Day 2 alongside Nasdaq's Michael Blaugrund and Jake Chervinsky.
Two data points anchored his appearances. First, HIP-3 — the permissionless market-creation framework launched in October 2025 — accounted for 51% of Hyperliquid's trading volume as of July 2026, with perpetual contracts on crude oil and pre-IPO equities as lead applications, according to KuCoin. Second, Hyperliquid's 30-day protocol revenue stood at $72 million at the time of the conference, making it the third-highest-revenue protocol tracked by DefiLlama, according to KuCoin.
Yan announced that options trading will be Hyperliquid's next major product, enabling hedging against spot and perpetual positions on shared order books, according to Phemex. No launch date was specified.
On the Day 2 panel, Yan critiqued the traditional finance model: "certain assets are tradable by only a select few during most of their growth phase, and by the time the public gains access, the gains have already been captured by a privileged few; this wealth creation model is unsustainable," per KuCoin. He positioned Hyperliquid as open infrastructure, noting that Builder Codes allow developers to access its markets and liquidity without building proprietary systems.
Polymarket CEO Shayne Coplan delivered a fireside chat on Day 1. The headline was his disclosure that Intercontinental Exchange (ICE), parent company of the NYSE, is one of Polymarket's largest shareholders and is exploring on-chain equity, per KuCoin. Coplan described an on-chain asset with "real programmable utility" tied to the Polymarket economy, characterizing it as resembling an ideal design of stocks, according to Crypto Briefing. No token name, ticker, distribution structure, or timeline was provided.
Separately, Coplan outlined a broader thesis: prediction markets as financial infrastructure rather than betting platforms. "The moment you put a price on a question or a risk that had no price, that price becomes a reference point for much larger financial transactions," he said, according to Seoul Economic Daily. He noted that Polymarket's pricing data is already referenced in over-the-counter deals, swaps, and block trades.
Parts 2 and 4 of this series scored both announcements: Hyperliquid's revenue disclosure as "shipped," the options product as "announced," and Polymarket's on-chain asset as "narrative pre-positioning."
Data as of early October 2026, sourced from DefiLlama and TokenPost unless otherwise noted.
| Metric | Value | |--------|-------| | 30-day perp volume | ~$205B | | 30-day gross fees | $90.6M | | 30-day protocol revenue | $73.4M | | Annualized fee run-rate | ~$929M | | Annualized revenue run-rate | ~$707M | | Open interest (total) | ~$17.3B | | Share of global perp OI | 11.9% (record) | | Open positions | >435,000 | | Active traders | ~320,600 | | Market share (perp DEX, 30d) | 31.9% | | RWA perp OI (July record) | $3.6B | | HIP-3 share of volume | ~51% (July 2026) | | HYPE cumulative buybacks | >$1.16B | | HYPE tokens retired | ~44.5M |
Quarterly revenue trajectory tells a more nuanced story than the headline figures. Gross protocol revenue peaked at $357 million in Q3 2025, then declined for four consecutive quarters: $295 million (Q4 2025), $217 million (Q1 2026), $202 million (Q2 2026), before a partial recovery to $224 million in Q3 2026, according to CoinPaprika. The 43% decline from peak to Q2 2026 trough occurred while trading volume and open interest hit records — a counterintuitive result explained by the HIP-3 fee-sharing structure detailed below.
Data as of early October 2026, sourced from DefiLlama, Sacra, and CNBC unless otherwise noted.
| Metric | Value | |--------|-------| | 2026 cumulative volume (through Oct 2) | $72.1B | | 30-day volume (Sep 2–Oct 1) | ~$4.6B in contracts | | 30-day gross fees | $66.4M | | 30-day protocol revenue | $16.4M | | Annualized revenue (June 2026 peak) | >$1B | | Fee rollout date | Jan 2026 (crypto); Mar 30, 2026 (broad) | | Fee structure | Taker-only: 0.04–0.07 by category | | Volume by category (approx.) | Sports ~39%, Politics ~32%, Crypto ~20% | | QCEX acquisition (U.S. reentry) | $112M (July 2025) | | CFTC designation | Amended Order, Nov 2025 |
Polymarket's revenue arc is the inverse of Hyperliquid's: from zero fees through all of 2025 to $1 billion annualized by June 2026. Five years of zero-fee operation was a deliberate liquidity subsidy, funded by venture capital, that built the volume base. When taker fees rolled out in January 2026 on crypto markets — expanding to sports in February and all major categories on March 30 — daily fee revenue jumped from $560,000 to over $1 million overnight, according to FinanceFeeds. The U.S. exchange launch in May 2026, following the QCEX acquisition for $112 million, per CoinTelegraph, pushed daily U.S. volume from $50 million to over $200 million within six weeks.
Most crypto protocols operate on a subsidy model: token emissions compensate liquidity providers, validators, and users at a cost that exceeds fee revenue. The protocol burns capital (via inflation) to acquire activity. Measured against this baseline, Hyperliquid and Polymarket are structural exceptions.
Hyperliquid charges trading fees on perpetual futures — a product with intrinsic demand from speculators, hedgers, and (increasingly) RWA traders seeking 24/7 exposure to oil, gold, and equities. Perps do not require liquidity mining incentives because the product itself attracts traders willing to pay for leverage and continuous settlement. The platform's cost of revenue — what it pays HLP vault depositors and HIP-3 builders — comes from fee revenue, not token emissions. Approximately 97–99% of remaining protocol fees flow to the Assistance Fund, which buys HYPE on the open market, according to Crypto.news. This creates a direct link between trading activity and token value — a fee-funded buyback rather than an inflation-funded subsidy.
Polymarket charges taker fees on prediction market trades. The fee formula — shares × rate × price × (1 − price) — peaks at 50-cent probabilities and declines toward extremes, generating 1.0% to 1.75% effective fees at the midpoint depending on category, per KuCoin. Maker fees are zero; a portion of taker revenue funds maker rebates to maintain tight spreads. The demand driver is information — traders pay to express views on elections, sports, economics, and geopolitics. This is not subsidized activity; users pay because the product (a binary contract with a real-world settlement trigger) is useful.
The distinction from the subsidy economy: neither protocol needs to inflate a token supply to attract users. Both charge fees because users value the product enough to pay.
Hyperliquid's headline revenue masks a structural shift. HIP-3, launched in October 2025, allows anyone who stakes 500,000 HYPE (approximately $27 million at current prices) to deploy a new perpetual futures market on Hyperliquid's order books. Builders keep up to half of the trading fees their markets generate, according to CoinPaprika.
The trade-off is visible in the numbers. HIP-3 markets grew from approximately 2% of perpetual volume in early 2026 to 51% by July. Over the same period, Hyperliquid's cost of revenue — the share of gross fees paid to builders — rose from under 6% (Q2 2025) to 18% (Q2 2026). Quarterly buybacks fell from $316.8 million (Q3 2025) to $192.3 million (Q1 2026), a 39% decline, per CoinPaprika.
This is an intentional design choice, not a failure. HIP-3 expanded Hyperliquid from a crypto-only perp DEX to a platform listing oil, gold, S&P 500, and pre-IPO equities — categories that drove the 51% volume share. Seven of Hyperliquid's top ten markets by volume in spring 2026 were tokenized equities or commodities, not crypto pairs, according to Crypto Briefing. The platform traded commodity perps, with WTI crude hitting $1.27 billion and Brent crude hitting $1.04 billion in single-day volumes in March 2026.
The economic question: does the growth in total fee volume from HIP-3 markets compensate for the reduced protocol take rate? Q3 2026 revenue of $224 million suggests a partial recovery from the Q2 trough of $202 million, but it remains 37% below the Q3 2025 peak. If RWA and equity perps continue scaling, the absolute revenue may recover. If the builder take rate compresses further or builders demand a higher share, the protocol's unit economics weaken.
Polymarket's regulatory position improved materially in 2025–2026. The CFTC issued an Amended Order of Designation in November 2025, permitting Polymarket to operate as a regulated, intermediated exchange in the United States, following the $112 million acquisition of QCEX, a CFTC-registered designated contract market, according to Crypto Briefing.
In January 2026, CFTC Chairman Michael Selig withdrew a Biden-era proposed rule that would have banned political and sports event contracts, per Sidley. A proposed rulemaking published June 10 would ban war and assassination markets while legalizing sports wagers — a framework broadly favorable to Polymarket's current product mix. Polymarket lifted its U.S. waitlist in May 2026 and reached $1 billion annualized revenue by June 26, per CNBC.
Residual risks: the CFTC's June proposed rule remains open for comment. Nine state-level legal challenges are pending. The regulatory framework for prediction markets is not settled; it is being written in real time.
Hyperliquid occupies a different position. Singapore's MAS stated in October 2026 that it is unaware of Hyperliquid being regulated in any major jurisdiction. Hyperliquid confirmed it is unregulated and has never claimed authorization from MAS or any equivalent body.
The path to U.S. access runs through a partnership with Bitnomial, a CFTC-registered exchange that would list regulated perpetual futures products mirroring Hyperliquid's markets — a proxy model, not direct access, per The Block. In August 2026, Hyperliquid's Policy Center urged the SEC and CFTC to harmonize rules for perpetual contracts.
Competitive pressure comes from above. CME and ICE executives have pushed the CFTC to scrutinize Hyperliquid over manipulation and sanctions-evasion risks, citing the platform's decentralized, no-KYC structure, according to CoinDesk. The CFTC cleared its first U.S.-regulated perpetual in May 2026; Kraken and Kalshi now trade regulated perps onshore, creating a compliant alternative that reduces Hyperliquid's relative advantage.
Real user fees do not automatically mean organic activity. Both platforms face volume-quality questions.
Hyperliquid. The HIP-3 builder model creates an incentive structure where builders earn up to 50% of trading fees from markets they deploy. Builders have economic motivation to drive volume to their markets — including through their own trading desks, affiliated market makers, or promotional incentive programs. Hyperliquid does not publicly disclose wash-trading detection methodology or volume-filtering data. The 11.9% share of global perp open interest and 435,000 open positions represent broad participation, but the distribution of volume across accounts is not transparent.
Polymarket. The maker rebate system, funded by taker fees, incentivizes liquidity provision. Polymarket published its fee formula and category rates but does not disclose the maker-rebate schedule in detail. The platform's volume concentration by category — sports at 39%, politics at 32% — raises the question of what happens to fee revenue outside major event cycles (elections, championships). The 2025 U.S. presidential election drove an outsized share of Polymarket's historical volume; the platform's ability to sustain post-election volumes was demonstrated through sports and crypto markets, but the June annualized peak coincided with the FIFA World Cup and the U.S. exchange launch — two tailwinds that may not persist simultaneously.
Neither platform has been subject to a public third-party audit of organic versus wash volume. This is not unique to these protocols — no major DEX or prediction market has undergone such an audit — but it means the fee-revenue figures should be read with the caveat that an unknown fraction of underlying volume may not reflect independent, arm's-length trading.
| Factor | Hyperliquid | Polymarket | |--------|-------------|------------| | 30-day gross fees | $90.6M | $66.4M | | 30-day protocol revenue | $73.4M | $16.4M | | Revenue retention rate | ~81% | ~25% | | Fee trend (quarterly) | Declining from Q3 2025 peak; Q3 2026 partial recovery | Rising from zero (2025) to $1B ann. (Jun 2026) | | Product demand driver | Leverage, RWA exposure, 24/7 trading | Information pricing, event outcomes | | Token subsidy required | No (fees fund buyback) | No (fees fund operations + maker rebates) | | Regulatory status | Unregulated; MAS/CFTC scrutiny | CFTC-designated; expanding U.S. access | | Revenue dilution risk | High (HIP-3 builder take rate rising) | Low (taker fee set by protocol) | | Concentration risk | RWA perps = 51% of volume | Sports + politics = 71% of volume | | Wash-trading transparency | No public methodology | No public methodology | | Next product | Options (no date) | On-chain asset (no date, no structure) |
Hyperliquid and Polymarket are exceptions in an industry that overwhelmingly subsidizes activity with token emissions. That distinction earned them prominent TOKEN2049 stage time and, more importantly, financial statements that TradFi participants in the next room could read without translating token-inflation math.
The exceptions come with caveats. Hyperliquid's fee-sharing model with HIP-3 builders is expanding the product surface (RWA perps, equities, commodities) while compressing the protocol's take rate — a classic platform trade-off between growth and unit economics. Polymarket's revenue surge coincided with multiple tailwinds (U.S. launch, World Cup, broad fee rollout) whose persistence is uncertain. Neither platform has resolved the wash-volume question that applies to every trading venue in the sector.
The economic-value test from Parts 1–4 of this series applies here with a qualification: real fees are necessary but not sufficient for durable economic models. Regulatory certainty, volume quality, and sustainable take rates determine whether these protocols remain revenue exceptions or revert to the industry mean.
Part 6 of this series will provide the full TOKEN2049 Week wrap-up and final scorecard.