Skip to main content
Leverage decides how much margin a position requires. At 10× a position of a given size ties up a tenth of its notional value; at 20×, a twentieth. What it does not do is change your exposure. A position’s profit and loss follow its size and the price, not the leverage you selected. Leverage changes how much collateral is committed and therefore how far price can move before that collateral is exhausted. Leverage is set per market, and unlike margin mode, it can be changed while you hold positions and live orders.

​
Maximum leverage falls with size

Every market has a tiered risk limit. As a position grows, the maximum leverage available to it drops and the maintenance margin requirement rises.
Values above illustrate the tier structure. Actual tiers, leverage caps, and rates are per market and published in each contract specification.
The logic is about unwinding. A small position can be closed into almost any book. A very large one cannot be closed quickly without moving the price against itself, which means the venue may not recover the full value when liquidating it. A higher maintenance requirement on large positions is the buffer for that.

​
Computing the requirement yourself

A tier is not just a rate. Each one carries four values, and the protocol resolves them into a requirement like this: IM=notional×im_leverage×10exponent\text{IM} = \text{notional} \times \text{im\_leverage} \times 10^{\text{exponent}} MM=max⁡ ⁣(notional×mm_leverage×10exponent−deduction,  0)\text{MM} = \max\!\left(\text{notional} \times \text{mm\_leverage} \times 10^{\text{exponent}} - \text{deduction},\; 0\right) The leverage figures are integers scaled by a shared exponent, which is how the same record yields both a maximum leverage and a maintenance rate. At im_leverage = 100, mm_leverage = 50, exponent = −4: initial margin is 1% of notional — hence 100× — and maintenance is 0.50%. The deduction is the term that matters and the one usually missing. Without it, a position that crosses a tier boundary would see its whole notional repriced at the higher rate, and its requirement would jump discontinuously. The deduction is set so that each tier’s line meets the previous one at the boundary: the higher rate applies, less a constant that cancels the step. The result is a requirement that rises continuously with size.
This is why a maintenance requirement computed from the rate column alone is wrong near a boundary — it will overstate the jump. Read maintenance_deduction along with the rate from the contract specification, or read the requirement the protocol reports for your position rather than deriving it.
One rounding rule is worth knowing: with a negative exponent the division rounds up. Margin requirements round in the protocol’s favour, never yours, so a requirement computed by hand may land a quantum below the real one.

​
Growing into a tier

The practical consequence: growing a position can move your liquidation price against you even if the price has not moved. Crossing into a higher tier raises the maintenance requirement on the whole position, which reduces the distance to liquidation.

​
Changing leverage with a position open

Raising leverage frees margin. Lowering it requires more, and is only permitted if you have enough.
Recomputed together
Request a leverage change
Initial margin on open positions
Margin reserved by resting orders
Closing fee held against each position
Still solvent?
Applied
Rejectedleverage unchanged
For an isolated position the same check runs inside that position’s own allocation.
The recomputation covers everything the change touches, not just the position: initial margin on open positions, margin reserved by resting orders, and the closing fee held back against each position. If the result would leave the account unable to cover its own commitments, the change is rejected and the previous leverage stands. For isolated positions the same check applies within the position’s own allocation, comparing what is committed against what the new leverage would require.

​
When a position outgrows its leverage

A position can end up above the tier its leverage allows — most often because it grew through fills, or because the market’s limits were tightened. The protocol does not liquidate for this, and it does not cancel what you already have:
  • Existing orders stay. They are not cancelled for being out of tier.
  • New risk-increasing orders are rejected.
  • At matching, a fill that would increase risk is rejected and the order cancelled.
  • Reducing is always allowed.
You resolve it by lowering leverage or reducing exposure. Until then the position is frozen in size — it can shrink but not grow.

​
Choosing leverage

The number people focus on is the one that matters least. What matters is position size relative to your collateral, because that is what determines your liquidation distance. Selecting 50× and opening a small position is safer than selecting 5× and opening one ten times larger. Two things leverage does control directly:
  • How much collateral is committed, and therefore how much is left free for other positions or as buffer.
  • Your deleveraging ranking, which is computed from unrealized profit multiplied by effective leverage. Higher leverage puts you earlier in that queue.

​
Reading the live values

Every figure on this page is on-chain per-market configuration, not a constant of the protocol. The parameters that drive it: They are published in each contract specification and readable from the chain. An integration that hard-codes them will eventually compute a number the network is not applying — see Developers.

​
Where to go next

Margin modes

Cross and isolated, and what each means for shared collateral.

Add margin

Adjusting an isolated position’s allocation without changing leverage.

Liquidations

How the maintenance requirement becomes a liquidation trigger.

Markets

Per-market tier tables and leverage caps.