Curve reports 704 soft-liquidation cases as some borrowers stayed in danger zones for months

Curve reports 704 soft-liquidation cases as some borrowers stayed in danger zones for months

N
News Editor
2026-09-08 12:56:56
Curve Finance has disclosed 704 soft-liquidation cases so far in 2026, covering 602 borrower addresses and highlighting how its LLAMMA-based lending model behaves very differently from standard DeFi liquidation systems. The data shows a median duration of 14.5 days in soft liquidation, while 25% of cases lasted at least 38.9 days. Some positions remained in the liquidation band for months rather than being closed immediately. Unlike Aave or Compound, where collateral can be sold as soon as a liquidation threshold is breached, Curve uses a price-band mechanism that gradually converts collateral into the borrowed asset. If market prices recover before a full liquidation, the process can reverse and borrowers may recover part or even all of their collateral. The figures also show that soft liquidation still carries real costs. Borrowers face DEX swap fees, slippage, rebalancing losses, and ongoing interest while the loan stays open. Curve’s disclosure adds to a broader discussion across DeFi about whether liquidation design should rely on hard thresholds or more gradual risk-sharing models.

Curve Finance has disclosed 704 soft-liquidation cases so far in 2026, involving 602 borrower addresses, offering a detailed look at how borrowers can remain in a danger zone for weeks or even months without being forcibly closed out.

According to data released by Curve, the median time spent in soft liquidation was 14.5 days, while the upper quartile reached 38.9 days. Some positions stayed in the liquidation range for months. Of the 704 cases, 476 took place in the first half of 2026, or about 68% of the total.

LLAMMA replaces a single liquidation line with a price band

In major lending protocols such as Aave and Compound, liquidation is usually straightforward. Borrowers post ETH or other tokens as collateral, and if the collateral price falls below a preset threshold, part of that collateral is sold to repay the debt. Once the sale happens, the borrower does not recover those assets even if the market rebounds later.

Curve’s LLAMMA system changes that structure. Instead of relying on a single threshold that triggers an immediate liquidation, it uses a price band. When collateral enters the danger zone, the protocol gradually converts collateral into the asset that was borrowed. If the market recovers before a full liquidation is completed, that conversion can be reversed, allowing the borrower to recover part or even all of the collateral.

What the 704 cases show

Curve’s figures break down the pattern behind these soft-liquidation events:

  • 704 soft-liquidation cases involving 602 borrower addresses
  • Most borrowers experienced soft liquidation only once
  • Median duration of 14.5 days
  • 25% of cases lasted at least 38.9 days
  • Some positions remained in the liquidation range for months
  • 476 cases occurred in the first half of 2026, about 68% of the total

These borrowers were not sitting in a grace period. Their collateral had already been gradually converted as prices moved lower, while the loans themselves remained open. That left some positions partially liquidated for days or weeks, with the possibility of recovery if prices turned back up.

Soft liquidation keeps a position alive, but not for free

Curve’s data also makes clear that soft liquidation comes with costs. Borrowers still absorb DEX swap fees each time the protocol converts assets. They can also face slippage during sharp price swings, repeated rebalancing costs if the market moves back and forth, and ongoing interest because the debt remains outstanding.

Even when a market recovery allows a position to survive, the borrower usually does not return to the starting point. Part of the collateral may already have been converted, and fees have already been paid. If the market keeps falling, a soft-liquidation position can still move into hard liquidation and be forcibly closed.

Curve’s scale and LLAMMA’s place in the protocol

According to DefiLlama data cited in the report, Curve has about $1.35 billion in total value deposited. Its DEX handled about $3.4 billion in trading volume over the past 30 days, generating about $4.3 million in fees and $1.15 million in protocol revenue over the same period. Outstanding loans stood at about $46 million.

Within Curve’s lending system, LLAMMA is presented as one of the protocol’s main innovations in DeFi lending. Standard liquidation models are binary: a position is either safe or liquidated. LLAMMA turns that boundary into a buffer zone, giving borrowers a chance to stay alive through short-term volatility.

A broader shift in DeFi liquidation design

The 704 cases point to a wider change in how DeFi lending protocols are thinking about liquidation. Under traditional models, sharp market moves can force borrowers out of positions during brief price crashes even when their collateral would have been sufficient over a longer horizon. The report said this kind of flash liquidation was repeatedly seen during the 2022 LUNA collapse and during several nonfarm payroll release days in 2025.

LLAMMA offers another approach: using time as a buffer when a position enters the danger zone and giving borrowers a chance to wait for a rebound. The trade-off is that borrowers take on more trading costs and slippage risk instead of having losses realized all at once by the protocol or liquidators.

Whether other DeFi protocols will adopt similar models remains open. The report noted that Aave v4 has started introducing more flexible liquidation parameters, while Compound is also exploring progressive liquidation. If Curve’s data eventually shows that soft liquidation delivers better overall risk control than traditional designs, DeFi lending may be moving into a new phase of liquidation mechanics.

This article was originally published by Bit.Fan. For more cryptocurrency news and market insights, visit www.bit.fan.
1000

Disclaimer:

The market information, project data, and third-party content displayed on this platform are for industry information sharing only and do not constitute any form of investment advice or return commitment.

Cryptocurrency trading carries high risks. Users should fully assess their risk tolerance and make independent decisions. All profits, losses, and legal responsibilities are borne by the users themselves.