Polymarket Perps Review 2026

Polymarket Perps is a perpetual-futures venue that Polymarket launched publicly on 3 September 2026. The first thing to be precise about is what it is not: this is a separate product from Polymarket's prediction markets. The two share a brand and an operator, and nothing else that matters to a trader. Prediction markets settle binary outcomes against a resolution source; this venue runs perpetual contracts with a funding rate, an order book, mark prices and open interest, on the same mechanical model as every other perpetual venue ORBIT tracks. Conflating them will mislead you about how the product behaves.

What is Polymarket Perps?

The catalogue is 67 perpetual markets, and its shape is worth reading carefully: 24 crypto, 36 equity, 4 commodity and 3 index. That makes it a mixed venue whose majority is tokenized TradFi rather than crypto — an unusual balance. Most mixed venues bolt a handful of equity markets onto a crypto core; here the equity half is the larger half by market count, which puts it closer in character to Ondo or QFEX than to a conventional crypto DEX, even though the crypto majors are where the deepest books sit.

The collateral asset is pUSD, not USDC. That is a small detail with a practical consequence: your margin lives in an asset specific to this venue, and moving capital in or out of a delta-neutral position spanning Polymarket Perps and a USDC-settled venue involves one more conversion step than a USDC-to-USDC pairing would. It is not an obstacle, but it belongs in your operational plan rather than being discovered at the moment you want to rebalance.

Order placement is geographically restricted. The United States, Canada and a sanctions list are barred from trading. This gates trading, not data — every public endpoint answers an anonymous read, which is why ORBIT can track the venue's funding, depth and prices regardless of where the reader sits. But if you are in a restricted jurisdiction, the funding rates on this page are information about the market, not an opportunity available to you, and no amount of favourable spread changes that. Check eligibility before building anything around it.

This review covers what the venue is and how it differs from the prediction-market product, its live metrics and the one metric it refuses to publish, the fee ladder and why we quote only its base rung, the measured depth across all 67 markets, the funding conventions including two traps that nearly shipped wrong, the eight markets that quote funding against the cross-venue consensus and why that is not a defect, and how it sits against every other venue tracked on ORBIT.

Polymarket Perps key metrics (2026)

The figures below are read live from ORBIT rather than written into the article, so they cannot go stale. One field will be conspicuously absent and the absence is deliberate: this venue does not publish 24-hour volume at all. Its ticker endpoint carries no turnover field, so rather than print a fabricated zero — which readers reasonably interpret as "nobody trades here" — ORBIT leaves it undefined. The honest consequence is worth stating: our automated open-interest plausibility monitor keys on the ratio of OI to turnover, so it structurally cannot run on this venue. Open interest was instead sanity-checked by hand against book depth, and the result was reassuring rather than alarming: BTC showed roughly $10.15M of open interest against $3.36M of resting asks, a ratio near 3x that sits squarely in the normal range for a functioning perpetual market.

Polymarket Perps key features for traders

The feature that most distinguishes this venue from other recent launches is not visible on its front page: it arrived with months of funding history already accumulated. The public history endpoint serves 2607 hourly data points reaching back to 7 May 2026 — roughly four months before the public launch date. The venue was evidently running privately for that whole period. The practical consequence is unusual and genuinely valuable: the backtester works properly from day one. Most new venues launch into an empty history, and ORBIT has to accumulate rates organically for weeks before a 30-day backtest means anything — SoDEX, Bullet and Perpl each went through exactly that. Here you can run a real historical funding backtest the day you discover the venue.

The second is the depth, which was measured across the entire catalogue rather than sampled on the majors. Of 67 markets, 66 fill a $10k order at 100%, 64 fill $100k, and 40 fill $1M; the deepest books hold between $2.7M and $3.9M. For a venue four weeks into public operation that is a serious liquidity profile, and the fact that 40 markets absorb a million dollars puts most of the catalogue genuinely in range for a sized funding position rather than merely being quotable.

• 67 markets, majority tokenized TradFi: 24 crypto, 36 equity, 4 commodity, 3 index — an unusual balance where the equity half outnumbers the crypto half.
• Months of funding history at launch: 2607 hourly points back to 7 May 2026, so the backtester is fully usable immediately rather than after weeks of accumulation.
• Hourly funding on a clean grid: history gaps are exactly 3600 seconds and the next-funding timestamp lands on the round hour.
• Depth measured across the whole catalogue: 66 of 67 markets fill $10k completely, 64 fill $100k, 40 fill $1M.
• Prices agree with the cross-venue cluster: 66 of 67 markets sit within 0.56% of the consensus price across every other venue we track.
• Published fee ladder: the base tier is 4.0 bps taker / 1.25 bps maker, with higher tiers cutting taker to 2.0 bps and paying maker rebates.
• Single bulk endpoint: mark, index, last, mid, open interest, funding rate and next-funding time all arrive in one call, so a full ingest cycle costs a single request.
• Live-tracked on ORBIT: funding, depth and cross-venue spreads appear beside every other venue on the Funding Screener.

Polymarket Perps sign-up bonus & fee discount

There is no token and no points programme attached to this product, and that should be stated plainly because the parent brand invites the assumption otherwise. Polymarket as a company has a long history of user incentives on its prediction-market side, and it would be easy to assume the perpetuals venue carries something similar. Nothing of the kind has been announced for it. If you are evaluating this venue, evaluate it on funding, fees and depth — the things it actually offers today.

There is also no referral or sign-up bonus to report. ORBIT links to the venue directly rather than through an affiliate path, for the simple reason that no such path exists here. Where a referral link exists we use it and say so; where none exists, inventing an incentive to make a page look more attractive would be a lie about the product.

What genuinely functions as a discount is the fee ladder. The rates ORBIT quotes — 4.0 bps taker and 1.25 bps maker — are the $0 volume tier, the rung everybody starts on. The published ladder cuts the taker fee to 2.0 bps at higher volume and pays maker rebates at the top. ORBIT deliberately models the base tier only, under a standing rule that applies to every venue on the site: quoting a discounted tier makes every spread look more profitable than it is for the reader who has not earned that tier, and most readers have not. If your volume genuinely qualifies you for a better rung, your real costs are lower than anything shown here — which is the right direction for an estimate to be wrong in.

The eligibility question sits above all of this. Trading is geo-restricted — the United States, Canada and a sanctions list cannot place orders. No fee tier, funding rate or depth figure matters if you are in one of those jurisdictions, so that is the first thing to check rather than the last.

Polymarket Perps trading fees

Polymarket Perps charges 4 bps taker and 1.25 bps maker on perpetuals. On a round-trip — entry and exit, and across two venues if you trade delta-neutral — those fees are the first thing any spread has to overcome. ORBIT's backtester subtracts both legs' taker fees plus live order-book slippage, so the PnL it shows is net, not headline.

Two things about these numbers deserve unpacking. The first is provenance: 4.0 bps taker and 1.25 bps maker are the base rung of a published volume ladder, not a flat schedule. Higher tiers cut the taker fee to 2.0 bps and pay maker rebates, so a high-volume account genuinely pays less than this article assumes. ORBIT models the base tier on every venue by standing rule, because a screener that quietly assumes VIP pricing tells most of its readers a spread is profitable when it is not. The second is the shape of the ladder: a 1.25 bps maker fee is low, and the gap between taker and maker is wide enough that resting orders change the arithmetic materially on a venue with books this deep. For a delta-neutral pair, though, the number that decides profitability is the full round trip across both venues plus live slippage, and ORBIT's backtester subtracts both from real order-book depth rather than from a fee table.

Funding rates on Polymarket Perps

Polymarket Perps settles funding every 1h. Funding is the payment between longs and shorts that anchors the perpetual to spot — and because every venue computes its own rate, the same asset can pay very differently on Polymarket Perps than on another exchange at the same moment. That gap is a tradeable, delta-neutral edge.

Is Polymarket Perps safe?

The funding conventions on this venue are unusually well pinned down, and that is worth treating as a safety property in the sense that matters to a screener user: numbers you can trust not to be silently wrong by a factor of eight. The scale of the funding field was proven three independent ways rather than assumed. Their value of 1.25e-05 annualises to exactly 10.95%, which is the industry base rate of 0.01% per 8 hours that dozens of venues in our own database sit on. A second clustered value, 6.25e-06, is exactly half of the first — it is the base rate on their RWA markets. Two exact hits on round, industry-standard numbers are impossible under a wrong scale factor. For completeness, reading the field as a per-8-hour rate gives 1.37% and reading it as a percentage gives 0.11%, and neither matches anything in the cross-venue cluster.

A second convention nearly shipped wrong and the near-miss is instructive. The venue exposes two funding fields: the ticker endpoint carries the rate for the hour in progress, while the history endpoint carries rates already paid. They are different numbers. On BTC they agree exactly — because BTC sits pinned to the base rate 24 hours out of 24 — so a check performed on BTC alone would have "proved" they were the same field. That is precisely the mistake ORBIT made on another venue once before, and the fix was to measure on markets that actually move: ONDO showed −3.381e-05 against −4.12e-05, GOLD showed 1.194e-05 against 6.25e-06, and BNB showed +1.11e-05 against −5.995e-05 — an opposite sign. ORBIT publishes the predicted rate, the one for the hour in progress, because that is what a trader deciding whether to open a position right now needs.

The funding interval is parsed rather than hardcoded, and there is a genuine trap in it: the venue reports its interval as the string "1h", not a number. Reading that field naively as a number yields NaN, and a hand-written constant substituted in its place would rot silently the day the venue changed cadence — which has happened to ORBIT on three other venues. Here the parsed value is confirmed independently by the data itself: history gaps are exactly 3600 seconds and the next-funding timestamp lands on the round hour.

Open interest is reported in base units, which was checked rather than inherited from another venue's convention. BTC reports 127.49; read as dollars that would be $127 against a book holding $3.3M, which is absurd, while read as base units it is $10.15M, which is sane. The same unit trap has caught ORBIT before on a different venue, so it is now verified per integration rather than assumed.

The custody and jurisdiction picture is the part to weigh personally. Margin sits in pUSD on the venue's own rails, and order placement is barred from the United States, Canada and a sanctions list. That restriction is a compliance posture rather than a technical failing, but it is a hard constraint: for a restricted reader this venue is a data point about where funding sits across the market, not a place to trade.

Polymarket Perps risks and considerations

1. Trading is geo-restricted. The US, Canada and a sanctions list cannot place orders. Data is public to everyone; access is not. Check eligibility before anything else.
2. Volume is not published at all. The venue exposes no turnover field, so ORBIT shows no 24h volume rather than inventing a zero. The consequence is concrete: our automated OI-plausibility monitor keys on the OI-to-volume ratio and structurally cannot run here, so open interest on this venue has one fewer automated cross-check than on venues that publish both.
3. Eight markets quote funding against the cross-venue consensus. BNB, ONDO, NEAR, WLD, PUMP, KSHIB, NVDA and TAO carry the opposite sign to our 40-venue median. This is explained rather than unexplained (see below) and all eight fill $10k at 100%, but it is a property you should understand before trading those specific names.
4. The majority of the catalogue is tokenized TradFi. 36 of 67 markets are equity. Those reference a price feed and a wrapper rather than a share you hold, and outside US market hours their primary market is closed, so price discovery is thinner and more oracle-dependent than on the crypto book.
5. pUSD collateral adds an operational step. A delta-neutral position spanning this venue and a USDC-settled one involves an extra conversion when you move or rebalance capital.
6. Young venue. Public since September 2026. The private-period funding history is a real advantage, but it is not the same as four months of public, adversarial operation with outside capital at risk.

How to get started with Polymarket Perps

1. Check eligibility first — order placement is barred from the US, Canada and a sanctions list, and that determines whether anything else on this page is actionable for you.
2. Be clear which product you are opening: Polymarket Perps is the perpetual-futures venue, distinct from Polymarket's prediction markets.
3. Fund margin in pUSD — note the collateral asset differs from the USDC most venues use, which matters when you pair this venue with another.
4. Open the Funding Screener and find where Polymarket Perps diverges from another venue on the same asset. On a venue whose majority is tokenized equity, check the crypto and TradFi halves separately — they behave differently.
5. Verify the specific pair absorbs your size on Size & Depth. Depth measured well across the catalogue, but 40 of 67 markets clearing $1M means 27 do not.
6. Confirm the net edge in the backtester — and take advantage of the fact that, unlike most new venues, the funding history here is deep enough for a 30-day backtest to mean something on day one.

Polymarket Perps vs Hyperliquid

The most useful comparison is Hyperliquid, the deepest on-chain perpetual venue and the default anchor leg for most DEX funding pairs — it is the benchmark against which a new venue's depth and funding behaviour are most readable. The second, and arguably the more apt one, is Ondo, because Ondo is the other venue on ORBIT built around a mixed crypto and tokenized-TradFi catalogue. That comparison tells you more about how this venue behaves in practice than a pure-crypto benchmark does: the equity half of the book has its own depth profile, its own session-hour effects and its own oracle dependency. For a delta-neutral trade the question is never which venue is better in the abstract — it is which one is currently paying the wrong funding relative to the other, and whether both books hold your size.

Polymarket Perps is also frequently weighed against Ondo — see the Polymarket Perps vs Ondo comparison for the full breakdown.

Polymarket Perps review: verdict

Polymarket Perps is a more serious venue than its four weeks of public operation suggest, and the reason is that it was not really four weeks old when it launched: months of hourly funding history were already on file, which makes it the rare new venue where a historical backtest is meaningful immediately rather than after a month of waiting. Depth was measured across the whole catalogue rather than sampled — 66 of 67 markets fill $10k completely, 40 clear a million — and prices agree with the cross-venue cluster on 66 of 67 markets within 0.56%. Those are the marks of a venue built properly rather than shipped fast. The honest limits are specific: trading is barred from several major jurisdictions, volume is not published at all and that costs the venue one layer of automated validation on its open interest, the majority of the catalogue is tokenized equity with the oracle dependency that implies, and eight markets quote funding against the cross-venue consensus for reasons that are explained but still worth understanding before you trade them. If you are eligible, it belongs in the rotation on measured depth and real funding history alone — and the fee ladder, whose base rung is what this article assumes, only gets cheaper from there.