# DeFiLoops and DeFi Saver do different things

> The closest thing to a competitor we have, and on most measures it is ahead. Here is what each is actually built for, with the numbers.

- Source: https://defiloops.com/blog/defiloops-and-defi-saver-do-different-things
- Published: 2026-06-16
- Category: Architecture
- Tags: comparison, automation, defi-saver, execution
- Author: DeFiLoops

---
If you are automating DeFi positions today, [DeFi Saver](https://defisaver.com/) is probably
the tool you have already heard of, and it is the closest thing to a competitor we have. It
is also, on most of the measures people compare on, ahead of us.

This page is the comparison we would want if we were the one evaluating.

## Where they are ahead, plainly

<Spec title="DeFi Saver, from their own published material" rows={[
  ['Chains', 'Five: Ethereum, Arbitrum, OP Mainnet, Base and Hyperliquid L1 — against our three'],
  ['Lending protocols', 'Aave v3/v4, Compound, Maker, Spark, Morpho, Fluid, Sky, Liquity, CurveUSD, LlamaLend, Euler v2 — against our Aave v3, Aave v4 and Morpho'],
  ['Maturity', 'Around $322m in total value locked, and an automation record through real mass-liquidation events'],
  ['Liquidation protection', 'Automated 24/7 monitoring that adjusts a position when it approaches a dangerous ratio. We do not have an equivalent'],
]} />

That last row is the one to weigh hardest. If what you want is a leveraged position that
defends itself against liquidation, that is what DeFi Saver is *for*, it has been doing it
for years, and it has been tested by events rather than by backtests.

<Callout type="warn" title="We do not have automated liquidation protection">
  Our schedules and price triggers can re-run a workflow, but there is no equivalent of a
  monitor that watches a health factor continuously and unwinds for you. If that is the
  requirement, this is not a close call.
</Callout>

## One assumption we got wrong, and are correcting in public

Our first draft of this page argued that DeFi Saver has you assemble steps by hand while we
compose from a sentence.

That is false. Their Recipe Creator supports natural language — you describe what you want
and it prepares the recipe steps. We checked before publishing, which is the only reason
this paragraph is an anecdote rather than a claim.

## What is actually different

Three things, and they are narrower than a feature table suggests.

### 1. A recipe is one transaction; a plan can span chains and wait

A Recipe is a set of actions combined into **a single transaction**. That is a real strength
— atomicity, no half-finished state — and it is also a boundary: a single transaction cannot
contain a fifteen-minute wait.

Our plans are built around exactly that wait. A crossing means value leaves one chain, time
passes, and value arrives on another, so a plan is a sequence of transactions with a gap in
the middle, and the progress lives in a database so a run survives us shipping an update
mid-flight.

That boundary is not something DeFi Saver chose, and it is worth saying so on their behalf.
Flow's [DeFiActions](https://github.com/onflow/FlowActions) framework — a separate project, a
different chain, a different language, and no competitor of ours — writes the same constraint
into its own README: every component "execute[s] within single transactions, ensuring that
complex multi-step financial operations either complete entirely or fail safely without
partial execution."

Two unrelated designs arriving at the same line is the clue. It is a property of what a
transaction *is*, not a limitation either team settled for. Which is also why the alternative
costs what it costs: to hold a fifteen-minute wait you give up atomicity, and then you owe the
reader an honest account of what a half-finished plan looks like.

<Compare left="One transaction" right="A plan across chains" verdict>
  <Fragment slot="left">
    Atomic: all of it happens or none does. Cannot contain a wait, so it lives on one chain.
  </Fragment>
  <Fragment slot="right">
    Spans chains, with the crossing's wait built in. Not atomic end to end — it breaks into
    transactions, and the board shows you where before you approve.
  </Fragment>
</Compare>

Neither is better. They answer different questions, and if your strategy fits in one
transaction the atomic version is the better tool.

### 2. Automation on a condition, against a schedule that re-prices

Their automation fires when an on-chain condition is met — a health factor crossing a
threshold, for instance. That is the right shape for protection.

Ours fires on a clock, a price, or another workflow finishing, and **rebuilds the plan each
time**. A schedule set up in March does not carry March's prices into June. That is the right
shape for a recurring strategy and the wrong shape for liquidation defence, where you want a
continuous monitor rather than a periodic rebuild.

### 3. The half that spends nothing

Five of our fifty-two tools move money. The other forty-seven only read: what a price range
would have collected day by day, how seventeen ways of managing it would have performed,
what nine option structures cost today and at every roll since 2021, what perpetuals pay
across four venues.

That is a research surface rather than an execution feature, and it is the part of what we
do that has no counterpart in a position manager.

## Fees, carefully

These are not directly comparable and anyone presenting them as a single number is being
sloppy.

<Spec rows={[
  ['DeFi Saver', '0.25% on advanced actions, plus 0.05% when those actions are automated. Standard supply, withdraw, borrow and repay are free'],
  ['DeFiLoops', '0.1% on a swap’s output, 0.1% on an amount sent between chains, and a markup on the gas bill charged in credits — not a cut of what a step moves'],
]} />

Different bases, different actions, different chains. The honest comparison is to price your
actual strategy on both, not to compare the percentages.

## Which to use

<Steps>
  <Step title="Pick DeFi Saver if you want a position that defends itself">
    Liquidation protection, on more chains and far more lending protocols, from a product
    with a real track record through liquidation events.
  </Step>
  <Step title="Pick DeFi Saver if your strategy fits one transaction">
    Atomicity is a genuine property and a recipe gives it to you.
  </Step>
  <Step title="Pick us if the strategy crosses chains and recurs">
    Plans built around the crossing's wait, re-priced every time they fire, that survive a
    restart mid-run.
  </Step>
  <Step title="Pick us if you want to measure before you commit">
    The read-only half is most of what we are, and it is free to use.
  </Step>
</Steps>

And the honest fifth option: use both. They are not the same tool, and nothing about holding
a position managed by one prevents you running a strategy through the other.

<Callout type="note" title="What we would want you to check on us">
  Three chains and nine protocols. No external audit. A guard that bans specific actions
  rather than permitting only known-good ones, with four attacks we have proven against
  ourselves. If those are disqualifying for you, they should be — and a comparison page that
  buried them would not be worth reading.
</Callout>