Probability Perpetuals · Arbitrum
Klærus
A perps protocol built on one idea: you should know exactly what you stand to win or lose before you enter a trade, and stay in control of it until you decide it ends. No leverage. No funding rate. No liquidation cascades.
See LiveThe problem
Every experienced crypto trader has lived the same moment: you call the direction, the market moves your way, and you still lose. Leverage amplified a wick. The funding rate drained the position overnight. A stop got hunted three seconds before the real move.
You were right. The platform still took what you were owed.
Traditional perps are engineered for institutions, and retail traders inherit machinery built for someone else's risk appetite. Strip a trade back to what a retail trader is actually deciding and only two questions remain: which direction, and over what timeframe. Everything layered on top of those two questions, margin, funding, liquidation price, is complexity that benefits the house.
That framing set the design constraint for the whole product: if a mechanic doesn't help the trader answer direction or timeframe, it shouldn't be in the interface.
Design approach
Klærus replaces the perps cockpit with a fixed-payout model. The trader picks an asset, a direction and a timeframe; the protocol prices that as a multiplier; the payout is known before any money moves. The design work was mostly about making that trade feel honest, and making the removal of familiar machinery read as clarity rather than as a missing feature.
Two modes instead of an order book
SWAP lets a trader back their own read of the market, pick Long or Short. FLIP assigns direction randomly at higher multipliers, for traders who want pure probability. Separating the two keeps a conviction trade and a coin-flip from sharing one ambiguous interface, and it makes the higher FLIP multiplier legible as compensation for giving up the choice.
Timeframe as the primary risk dial
With leverage gone, the timeframe ladder, 1 minute through 4 hours, becomes the main lever a trader pulls. Each rung carries its own multiplier, so the risk/reward trade-off is visible in the same control that sets it, rather than buried in a separate margin calculation.
A demo account with real price feeds
Probability perps are an unfamiliar instrument, and the fastest way to lose a curious trader is to make their first lesson an expensive one. Every account carries a permanent demo balance running on live market prices, with every mechanic available and a reset button. Learning the product costs nothing; only conviction costs money.
Showing the payout first
The single most important screen in the product isn't the chart, it's the calculator. Before entering, a trader sets a stake and a timeframe and sees not just the winning number but every outcome, including the ones that hurt. A $100 stake at a 2.40× multiplier resolves like this:
| Outcome | What it means | Result |
|---|---|---|
| Win at expiry | Direction was right when the timeframe closed | +$133 |
| Cut profit early | Exited while ahead, before expiry | +$4 to +$110 |
| Liquidated mid-trade | Stopout triggered before expiry | −$80 |
| Wrong at expiry | Direction was wrong when the timeframe closed | −$100 |
Publishing the downside next to the upside is a deliberate trust move. Trading products usually lead with the win and bury the loss; here the four outcomes sit in one view at the same visual weight, because a trader who can see the worst case before entering has no reason to feel ambushed by it afterwards.
Control mid-trade
Binary options have a deserved reputation problem, and most of it comes down to one thing: once you're in, you're locked in until expiry, all-or-nothing. Klærus keeps the simplicity of a binary payout but removes the trap.
- Cut Profit, lock in gains early rather than watching a winning position round-trip.
- Stopout Loss, cap the damage while the trade is still open.
- Breakeven Exit, walk away flat when conviction changes.
Designing these as three distinct, always-visible actions, rather than one generic "close position", was intentional. Each maps to a different emotional state a trader is actually in, and naming the state makes the right button easier to find under pressure.
The interface
Everything settles on Arbitrum, in USDT, from a non-custodial wallet. Deposits route through nQ-Swap so a trader can fund from any chain without a manual bridge step, the kind of infrastructure detail that decides whether someone completes their first trade.
Where it stands
Klærus is in closed beta on Arbitrum. More than 2,000 traders have requested access and roughly 5% have been let in, a cohort of about a hundred people who have put over $200,000 through the protocol.
Keeping the door mostly shut is the point rather than a bottleneck. A fixed-payout product has to be right about its multipliers before it scales, and a small cohort makes it possible to watch how real traders use the mid-trade exits, which timeframes they actually reach for, and where the interface still has to explain itself. The waitlist is a demand signal; the 5% is where the product gets corrected.
The demo account matters more than the funnel numbers suggest. It carries live prices and every mechanic, so the people arriving from that waitlist can learn an unfamiliar instrument before any of their own money is exposed.