Educational content, not investment advice. Crypto-asset values fluctuate.

Modulis 3 · DeFi lending
4/6

Aave.com — borrowing on the Ethereum network

Video · 2 min.
Aave: borrowing on Ethereum

Milly borrows against ETH on Aave. On 2026-03-10 a misconfigured price feed liquidated about $27M of positions while the market price stood still. How Aave differs from Kamino, which network to use, how Health Factor works, and why deposits are not always withdrawable.

Narrated, with light music
The full lesson with numbers and rules is below.
Watch on YouTube
ETH Aave

Aave is the largest DeFi borrowing platform in the world. People have trusted it with more than $15B. It's been running since 2020 and is the most battle-tested protocol in all of DeFi.

In our strategies, Aave is a backup platform — if something happened to Solana, Aave keeps running independently.

You'll see V3 and V4 when you open Aave — choose V3

Since 2026-03-30 there is also Aave V4, built differently inside ("Hub & Spoke"): funds sit in a shared hub, while separate modules set the collateral rules. This is not simply a newer version — it is a different risk structure.

As of 2026-09 the real market is still V3: it holds ~$17.4B versus V4's ~$0.38B (about 2%). This lesson works with V3. Choose V4 only once you understand how its structure differs.

Data as of 2026-09-05; the split is moving fast — check app.aave.com.

Once you understand Aave, you can build a position on two independent networks at once — no longer hostage to a single ecosystem, but someone who chooses where and when to act.

Why do you need both Kamino AND Aave?

Don't put all your eggs in one basket

Imagine you have two bank accounts. If one bank goes down, the other still works. Same with DeFi: if the Solana network runs into trouble, Aave on Ethereum keeps running on its own.

ReasonExplanation
BackupIf Solana goes down — Aave keeps running
A different assetETH moves differently than SOL
More liquidityAave TVL $18.7B — about 14x more than Kamino ($1.33B); 2026-09 figures, they move
Longer track recordRunning since 2020 — the most battle-tested
E-ModeA special mode with a higher LTV
Diversification

Never concentrate everything in one place. If 70% sits on Kamino and 30% on Aave — even if one platform runs into problems, the other portion stays safe.

Which network to choose for an Aave position

ETH Aave runs on several networks. The steps in this lesson use Arbitrum, but in 2026 that is no longer automatically the best choice — and it matters that you know why.

The old argument has stopped holding

We used to say it simply: "mainnet is expensive, go to an L2." That is no longer true. On 2026-09-05 an Ethereum mainnet swap cost about $0.04 (network gas ~0.05 gwei) — cheapness is no longer a reason to go anywhere.

Liquidity has moved too: Arbitrum's DeFi TVL fell 52% across 2026 (from $2.90B to $1.40B), while Base grew to $5.63B — roughly 4× the size of Arbitrum today.

NetworkDeFi TVL (2026-09)When it fits
Ethereum mainnet~$49.2BDeepest liquidity, no bridge. Fits if you act rarely and in larger amounts
Base~$5.63BThe largest L2 — if you want L2 speed and cost
Arbitrum~$1.40BStill works, but has the thinnest liquidity of the three
A bridge is not free

To reach any L2 you move assets across a bridge, and a bridge is a separate risk — not just a fee. The most expensive DeFi incidents of 2026 came precisely from bridges and their configuration.

In practice: if you plan one position and infrequent actions, mainnet with no bridge may be both simpler and safer. Choose an L2 for where the liquidity you need actually is — not because it is "cheaper".

Figures as of 2026-09-05 (etherscan.io/gastracker, DefiLlama). They change — check before choosing a network.

Arbitrum analogy

Ethereum mainnet is like a highway: usually clear and cheap, but during rush hour the toll can jump several times over. Arbitrum is a separate tunnel to the same destination, where the price barely changes no matter what's happening on the highway. The steps below use Arbitrum as the example, but the process is identical on every network — just switch networks at the top of app.aave.com.

Fee comparison

ActionEthereum mainnetArbitrum
Deposit ETH~$0.03–0.30~$0.01–0.05
Borrow USDC~$0.03–0.30~$0.01–0.05
Repay debt~$0.03–0.30~$0.01–0.05
An active week~$0.20–1.50a few cents
What an L2 actually gives you

After the 2024 "Dencun" upgrade, Ethereum mainnet fees dropped to cents — cost is no longer a barrier, and that is the only thing you need to remember about fees. What remains is predictability: mainnet fees move with network load and can briefly spike several times over during congestion, while an L2's stay nearly constant.

But predictability alone is not reason enough to pick an L2 — you pay for it with bridge risk and thinner liquidity. Before a larger operation on mainnet, check the current price at etherscan.io/gastracker (on 2026-09-05 a swap cost ~$0.04).

How do you get onto Arbitrum?

  1. Hold ETH in your MetaMask wallet

    The ETH needs to be on the Ethereum network in your MetaMask wallet.

  2. Move ETH to Arbitrum via a bridge

    Go to bridge.arbitrum.io. Move ETH from Ethereum to Arbitrum. Takes 10–15 minutes.

    The bridge transaction costs an Ethereum mainnet fee — usually ~$0.03–0.30, more during congestion.

  3. Add the Arbitrum network to MetaMask

    MetaMask → Settings → Networks → Add Network → Arbitrum One. Or once you connect to Aave, MetaMask will offer to add it automatically.

  4. Connect to Aave on Arbitrum

    Go to app.aave.com. Select the 'Arbitrum' network at the top. Connect with MetaMask.

ETH collateral on Aave — step by step

  1. Connect to app.aave.com with MetaMask

    Select the Arbitrum network. You'll see all markets with their interest rates.

  2. Deposit ETH as collateral

    Click 'Supply' on ETH or WETH. Enter the amount. Confirm in your MetaMask wallet.

    Aave uses WETH (Wrapped ETH — a technical wrapper; the conversion happens automatically).

  3. Turn on collateral mode

    Toggle the 'Collateral' switch on ETH — this lets it be used as collateral.

  4. Borrow USDC

    Click 'Borrow' on USDC. The rate will be Variable — Aave V3 doesn't offer another option anymore. Watch your Health Factor.

    Health Factor needs to stay above 1.5. If it drops below 1.0 — liquidation.

  5. Monitor your position

    The dashboard shows Net Worth, Health Factor, and your positions. Check it daily.

E-Mode — a special mode

Aave has a unique mode — E-Mode (efficiency mode). It lets you get a higher LTV when you borrow and post similar-type assets as collateral.

E-Mode analogy

It's like a bank saying: "If you want a euro loan and post euro bonds as collateral, I can lend you more, because both assets are very similar and barely move relative to each other."

ParameterNormal modeE-Mode
CollateralETHstETH (staked ETH)
DebtUSDC (different class)WETH (same class)
Max LTV80.5%93%
Liquidation penalty5%1%

Values as of 2026-09-05 (Aave V3 WETH market). Aave governance changes them — check app.aave.com.

E-Mode isn't used in our main strategies — we borrow USDC (a different asset class). But it's useful to know.

Kamino vs Aave — comparison

Kamino (Solana)Aave (Arbitrum)
NetworkSolanaEthereum (Arbitrum)
Main assetSOLETH
Speed~0.4 seconds~2 seconds
Fees~$0.01~$0.01–0.05
Trust (TVL)$2B+$15B+
Running since20232020
Collateral yield~6% (SOL)~2–3% (ETH)
USDC borrow cost~5–7%~5–7%
WalletJupiter WalletMetaMask
Network stabilityOccasionally goes downVery stable
Recommendation

Variable vs Stable Rate — which to choose?

ModeVariableStable (fixed)
InterestChanges every minuteWas temporarily fixed
APR~5–7%historically ~8–10%
Available on Aave V3YesNo — disabled since 2023

Choose Variable Rate — in practice it's the only option: Aave V3 disabled Stable-rate borrowing across all markets in 2023, so you won't even see it in the interface. And back when it did work, over the long run it tended to be more expensive than the variable rate.

Bonus: AAVE token rewards

Using Aave, you occasionally earn extra AAVE tokens as a reward (depends on active programs). This can lower your effective borrowing cost — the amount varies and isn't always guaranteed:

Aave risks — and how you manage each one

AUDITED Aave has 10+ independent security audits. Every risk here has a concrete lever — you don't fear it, you know how to keep it in the green zone.

RiskLevelHow to manage
Smart contract bugVery low10+ audits, running since 2020
Price feed (oracle) errorMEDIUMLiquidation is based not on the market price but on the one the protocol's price feed reports. On 2026-03-10 a misconfigured Aave price feed liquidated ~$27M of positions even though the market price had not moved. Management: keep extra headroom — never plan right up to the limit
Can't withdrawMediumYour deposit can be temporarily unwithdrawable if others have borrowed it. In April 2026 Aave WETH liquidity fell from $689M to $1.5M in about 2 hours. Management: don't keep money on Aave that you might need the same day
LiquidationManagedKeep Health Factor > 1.5
Arbitrum bridge riskMedium7-day protection period; the bridge is a separate risk. Management: if your actions are infrequent, consider mainnet with no bridge
Interest rate spikeMediumMonitor and refinance if > 10%
"Aave = safe" — with one caveat

Aave really is the most battle-tested lending protocol there is. But in 2026 its depositors took real losses — and not because anyone broke Aave's code. In April 2026, during an incident involving third-party collateral, the price feed kept reporting a stale price, so liquidations simply did not fire in time, and hundreds of millions of dollars of bad debt were left in the protocol.

What to take from this: liquidation is not purely arithmetic on the market price. It also depends on the protocol's price feed and on other users' collateral types, none of which you control. Headroom in your Health Factor is therefore not excessive caution — it is the price of what you cannot foresee.

In short

Quick check
You are choosing a network for an Aave position. Which statement is true in 2026?
Practice task
0 / 5
Explore the Aave market on Arbitrum

Now compare the two platforms yourself: open Aave and find the same numbers you saw in the Kamino lesson. Takes about 10 minutes, no wallet needed.

Just exploring — you're not depositing, borrowing, or confirming any transactions. Double-check that the domain really is app.aave.com — fake sites with similar addresses are a common scam.

This is learning, not investing — use only small amounts you treat as tuition.

You just read the Aave market on Arbitrum yourself and compared it with Kamino — now you can choose a platform based on numbers, not gut feeling.

Kamino.com (SOL)Borrowing strategy and looping
Mark as read

Support

Free material is easy to put off

Something practice teaches: when you paid nothing to learn, it is easy to postpone learning indefinitely. If these lessons actually gave you something — saved you a mistake, or explained what nobody else did — you can pay whatever you think they were worth.

This is not a price, a subscription, or a fee for access. The amount is yours to choose, and paying nothing is entirely fine.

Solana address

CUj6wYQLSBxidzm7Lqtkxbw2d8Sv2H2tbJELyAjR756M

Solana network only

Send SOL or SPL tokens (e.g. USDC) only on the Solana network. Sending from an exchange over a different network loses the funds permanently — nobody can recover them.

What you get for it

Nothing extra, and that is deliberate. Every lesson, all three difficulty levels and every calculator stay free and sign-up-free, with or without a donation. There is no gated section, no share of profits, no future token and no claim on anything.

Where the money goes

A donation becomes my own money. I use it to keep the site running and to fund a public demonstration position on a lending protocol — real money, so the lessons have something to show instead of only theory. I publish the results, losses included, publicly and free for everyone. It is my position and my risk, not a pooled fund.

This is a gift for content already received, not an investment and not a payment for a service. Non-refundable. Educational content, not an investment recommendation.