Simple Base SwapSimple Base SwapOpen app
← All articles
Nov 19, 2026·5 min read

Stableswap AMMs Explained: Why Stablecoin Pools Use a Different Curve

basedefiguides
base

If you have ever swapped one stablecoin for another on Base and noticed the price barely moved, even on a fairly large trade, you were not looking at a fluke. You were looking at a different kind of automated market maker than the one most swap guides describe.

Most explanations of how a decentralized exchange works start with the constant product formula, often written as x times y equals k. It is a simple, elegant idea: a pool holds two assets, and the product of their quantities stays constant as trades happen. It works well for assets whose relative price can move freely, like ETH and a random token. But it is a poor fit for two assets that are supposed to always be worth close to the same thing, such as two different wrapped versions of the dollar. Pools built specifically for that situation use a design usually called stableswap.

What a constant product pool does to similar assets

In a constant product pool, the price you get shifts noticeably as a trade gets larger, even when both assets are pegged to the same value. That happens because the formula does not know the two assets are supposed to track each other. It treats USDC and a second dollar token exactly like it would treat ETH and a meme coin: as a trade pushes the balance of the pool further from fifty fifty, the price it quotes moves further away from one to one. For a deep dive into that general mechanism, see our guide on price impact versus slippage.

For two tokens that should trade at parity, that behavior is wasteful. It makes a mechanically simple swap more expensive than it needs to be, and it gives arbitrage traders an easy, repeated opportunity to profit off ordinary users at the pool's expense.

Blending two formulas

Stableswap pools, a design popularized by Curve and since adopted in various forms across many chains, solve this by blending two different pricing curves instead of using just one.

The first is the constant sum curve, where the two balances simply add up to a fixed total. This curve gives near zero slippage, but on its own it has a serious flaw: if one asset in the pool ever becomes worth meaningfully more or less than the other, traders can drain the pool of the valuable asset at an unfair rate, since the formula never adjusts price based on how imbalanced the pool has become.

The second is the familiar constant product curve, which is self correcting but, as above, moves price more than necessary for assets that should stay near parity.

A stableswap pool runs a formula that behaves like the flat, low slippage constant sum curve while the pool is close to balanced, since that is the normal, expected state for two pegged assets. As the pool gets pushed further out of balance, usually because one asset is being bought up faster than it is being supplied, the formula gradually bends toward the more defensive constant product behavior. A setting usually called the amplification coefficient controls how wide that flat, low slippage zone is before the curve starts bending. A higher amplification coefficient keeps prices flatter across a wider range of pool imbalance, which is appropriate when the two assets are extremely unlikely to diverge in value, such as two forms of the same stablecoin.

Where you will actually run into this on Base

Decentralized exchanges on Base generally offer two different kinds of liquidity pools and let a token pair use whichever fits. Pairs expected to trade near a fixed ratio, like stablecoin to stablecoin, or a liquid staking token to the asset it represents, are usually routed through a stable pool using this blended design. Pairs with no fixed relationship, like a stablecoin paired against a volatile token, use a standard constant product pool, or the concentrated liquidity style pools covered in our guide on concentrated liquidity. A swap aggregator or router typically picks the right pool type for you automatically, which is one reason the same two tokens can sometimes be reached through more than one pool when you look at a block explorer.

The risk that this design does not remove

A stableswap pool is built around the assumption that both assets hold their peg. That assumption is the entire basis for offering such flat pricing in the first place, and it is also the design's main weak point. If one asset in the pool loses its peg, meaning it becomes genuinely worth less than the other asset rather than just temporarily imbalanced in the pool, the formula's flat pricing zone works against liquidity providers rather than for them. It will keep quoting the depegged asset closer to parity for longer than a simple constant product pool would, which means traders can offload the weaker asset into the pool at a better rate than its real market value, at the liquidity providers' expense. Our guide on stablecoin depegs covers what that situation looks like from a holder's side, and our guide on liquidity pools and impermanent loss covers the general risk of supplying liquidity.

The short version

Stableswap pools exist because a plain constant product formula is not an efficient way to trade two assets that are meant to be worth the same thing. By blending a flat, constant sum style curve with a self correcting constant product style curve, and controlling the balance between the two with an amplification coefficient, these pools offer much lower slippage for pegged assets while still having some defense built in if a pool becomes imbalanced. That defense has limits. It assumes the peg itself holds, and a real depeg is the one scenario where this design's main advantage, flat pricing, becomes a liability for the liquidity providers behind the pool.

Ready to try it yourself?

Create a non-custodial wallet on Base in seconds. No account, no sign-up.

Open the web app