To get into DeFi, you first need to fund an account on a centralized exchange, then withdraw funds to your own wallet. You can complete account setup and KYC through the Binance Official Site or the Official Binance App, and if you're on an Apple device that can't install directly, check the iOS Install Guide. DefiLlama is currently the most trusted DeFi data aggregator, tracking TVL (Total Value Locked) across 350+ chains and 3,500+ protocols. But TVL itself can be calculated several different ways, and the same protocol can show numbers that differ by up to 30% across different sites; the same chain's TVL can also treat asset types, bridged funds, and native-versus-wrapped assets very differently. Understanding TVL's limitations and nuances is essential before you use it to judge the scale of an ecosystem.
DefiLlama was founded in 2020 by 0xngmi (an anonymous developer), and in 2024 it raised $6 million from funds including GoldenTree. As of April 2026, DefiLlama tracks:
DefiLlama's core strengths:
Transparent data: all of the calculation logic is open source on GitHub, and the community can contribute new protocols and fix errors.
Completely free: unlike paid data platforms like Token Terminal or Messari, DefiLlama's core data has zero barrier to access.
Frequent updates: most protocols update every 5-15 minutes, pulled directly from on-chain contract reads.
Open API: developers can call the API for free (rate-limited, but sufficient for most use cases).
TVL literally means "the total value of assets locked in a protocol's contracts," but the calculation details get more complicated than that:
Direct TVL: the current dollar value of every ERC-20 token and native coin sitting in a contract. For example, if Aave's contracts hold $1 million in USDC plus 500 ETH, TVL = $1 million + 500 x the ETH price.
How borrowed funds are treated: in a lending protocol, deposits count as TVL and borrows count as negative TVL. Aave's "net TVL" equals total deposits minus total borrows. But DefiLlama shows gross TVL (total deposits) by default, since it's more intuitive.
How LP tokens are treated: an LP token represents a share of two underlying assets, valued based on the market value of those assets. Curve's 3pool TVL equals the total of the USDC, USDT, and DAI sitting in the pool.
How wrapped assets are treated: wrapped assets like WBTC and wstETH count toward TVL, but the underlying assets (BTC, ETH) are already counted in the source contract, so DefiLlama uses classification rules to avoid double-counting.
Traps that newcomers commonly fall into when reading TVL numbers:
A user deposits $1 million USDC into Aave, borrows $800,000 USDC, deposits that back in and borrows $640,000, and after looping this 5 times, the protocol shows $3.5 million in USDC TVL. But the user's actual principal is only $1 million.
This kind of recursive looping is common in leveraged farming and recursive lending. The more professional way to look at it is "unique deposits" (de-duplicated, real deposits), a metric DefiLlama provides for some protocols.
A user bridges $1 million USDC from Ethereum mainnet to Arbitrum:
The same dollar of funds can get counted across three separate TVL layers — the bridge, the chain, and the protocol. DefiLlama's "Chain TVL" and "Protocol TVL" are different dimensions, and it's important to be clear which one you're comparing.
Some protocols' TVL is heavily made up of their own token. For example, a new protocol's farm might require users to deposit its own token to farm, and end up with 90% of the protocol's TVL made up of that token. TVL like that is an entirely different tier from TVL made up of "hard currency" assets like USDC or ETH.
DefiLlama's "Stablecoin Flow" and "Token Diversity" metrics can help you judge this.
stETH (Lido's staking receipt token) counts toward Lido's TVL. If that same stETH gets deposited into Aave, it counts toward Aave's TVL too. If it's then restaked into EigenLayer, it counts toward EigenLayer's TVL as well. And if Aave's stETH is then supplied to a Curve LP, it gets counted again in Curve's TVL.
This kind of "nested counting" is extremely common across the ETH liquid-staking and restaking ecosystem, and it significantly inflates Ethereum's total TVL figure.
Beyond TVL, DefiLlama also provides:
Fees & Revenue: total protocol fee revenue and retained revenue. Uniswap generates roughly $80 million in monthly fees; Aave retains roughly $3.5 million a month at the protocol level.
Stablecoins: stablecoin circulating supply, cross-chain distribution, and mint/burn activity. USDT is 50% on Tron, 40% on Ethereum mainnet, and 10% spread across other chains.
Yields: a ranking of LP and lending APRs across every DeFi protocol, filterable by asset or chain.
DEX Volume: total DEX trading volume per chain.
Bridge Volume: total cross-chain bridge trading volume.
Treasuries: protocol treasury sizes (Uniswap's treasury holds $6 billion, Optimism's holds $2.5 billion, and so on).
Per DefiLlama data from April 2026:
| Rank | Chain | TVL | 24h Change | Main Protocols |
|---|---|---|---|---|
| 1 | Ethereum | $72B | +0.8% | Lido, Aave, MakerDAO |
| 2 | Solana | $14.5B | +2.1% | Jito, Marinade, Kamino |
| 3 | Tron | $10.5B | +0.3% | JustLend, SunSwap |
| 4 | BNB Chain | $7.8B | -0.5% | PancakeSwap, Venus |
| 5 | Base | $6.2B | +1.5% | Aerodrome, Aave, Morpho |
| 6 | Arbitrum | $5.8B | +0.7% | GMX, Aave, Camelot |
| 7 | Avalanche | $1.8B | +0.4% | Aave, BENQI, Trader Joe |
| 8 | Polygon | $1.2B | -0.2% | Aave, Quickswap, Balancer |
| 9 | Sui | $1.1B | +1.8% | Cetus, Navi, Suilend |
| 10 | Optimism | $0.85B | +0.5% | Velodrome, Aave |
Some notes:
TVL numbers look very different once you look at what's actually behind them on each chain:
The most mature DeFi ecosystem, characterized by:
The chain with the strongest DeFi revival in 2024-2026, characterized by:
Utility-focused, centered around USDT circulation:
Binance's ecosystem chain, characterized by:
Coinbase's Layer 2, the fastest-growing chain in 2024-2026:
Per DefiLlama's historical data:
| Date | Global TVL | Cycle Phase |
|---|---|---|
| December 2020 | $15B | DeFi Summer kicks off |
| November 2021 | $180B | All-time peak |
| December 2022 | $38B | Bear market low (post-FTX) |
| December 2023 | $70B | Recovery |
| December 2024 | $110B | Driven by RWA + LST growth |
| December 2025 | $128B | Continued growth |
| April 2026 | $130B | Current |
Global TVL has grown roughly 240% from the 2022 low to 2026, driven mainly by:
But it's still about 30% below the 2021 peak of $180 billion, which reflects that DeFi as a whole is still in a transitional zone between the last bear market and the next bull run.
DefiLlama is open source and free, but there are several complementary data sources worth knowing:
| Platform | Specialty | Free? |
|---|---|---|
| Token Terminal | Protocol financial data (P/E, revenue, etc.) | Partially paid |
| Dune Analytics | Custom SQL dashboards | Free |
| Messari | Industry research plus standardized data | Partially paid |
| Nansen | On-chain wallet labeling | Paid |
| Footprint Analytics | Multi-chain data visualization | Partially paid |
| L2Beat | Layer 2 data | Free |
DefiLlama is enough for most ordinary users; for research or development work, it's worth combining it with Dune (for custom queries) and Token Terminal (for deeper financial data).
A practical, step-by-step approach:
Step one: Check the TVL rankings. Under Categories, look at Lending, DEX, Yield, and other segments for the top protocols. Protocols with TVL over $100 million are generally lower-risk.
Step two: Check the TVL trend. Every protocol page has a 7d, 30d, and 180d trend chart. A steadily rising TVL combined with healthy user growth is a good sign; sharp swings in TVL or TVL concentrated in a handful of addresses is a warning sign.
Step three: Check the Yields ranking. The Yields tab lets you sort by asset, chain, or APY. Once you've found the asset you're interested in, look at the APR breakdown (base yield versus reward yield). If the reward portion is too high (over 50%), that annualized rate probably isn't sustainable.
Step four: Check whether Fees make sense. The Fees tab compares protocol fee revenue. "High TVL plus low fees" usually means capital isn't generating real transaction activity — a sign of farmable, artificial growth.
Step five: Check the protocol's security score. The "Audits" panel on the right side of every protocol page shows the auditing firms and links to reports. No audits, or only an anonymous audit, is a high-risk signal.
Q: Is DefiLlama's data actually accurate?
Generally, yes, but there are a few common sources of error: first, new protocols can take 1-3 days to get integrated; second, protocols with custom contracts sometimes aren't fully crawled; third, price feeds occasionally glitch and cause TVL to jump around. The community submits data-correction PRs on GitHub every day.
Q: Does high TVL mean a protocol is definitely safe?
Not necessarily. High TVL indicates user trust, but real safety also depends on code audits, how long the protocol has operated without incident, and whether it's ever been hacked. Anchor Protocol still had $18 billion in TVL right before it collapsed in 2022. It's best to weigh TVL alongside track record and audit history.
Q: How does DeFi's total TVL of $130 billion relate to Bitcoin's $2 trillion market cap?
DeFi TVL measures "funds locked in protocols," while Bitcoin's market cap measures "the total value of circulating BTC" — the two aren't directly comparable. BTC's market cap reflects store-of-value demand and speculation; DeFi TVL reflects "capital actively participating in on-chain finance." The latter is a better gauge of activity at the Web3 application layer.
Q: Is TVL correlated with a protocol's token price?
Only weakly. TVL growth doesn't necessarily drive token price (many protocol tokens have no cash flow rights); rising token prices don't necessarily drive TVL growth either. Uniswap's UNI price has been decoupled from its TVL for a long time, because UNI doesn't carry a claim on protocol fees.
Q: Why is there such a big gap between Solana's $14.5 billion TVL and Ethereum's $72 billion?
The main reason is that Ethereum holds a huge amount of LST/LRT assets (Lido alone accounts for $32 billion), while Solana's LST ecosystem is much smaller (Jito is roughly $5 billion). Strip out LSTs, and the gap in "native DeFi TVL" between the two chains narrows considerably (roughly $40 billion for Ethereum versus roughly $9.5 billion for Solana).
Q: How can you tell if TVL numbers are being artificially inflated?
A few signals to watch: first, the ratio of TVL to trading volume (a protocol with $100 million TVL but only $500,000 in daily volume is likely being farmed for show); second, the number of unique users (high TVL with very few users suggests domination by a handful of whales); third, how concentrated the protocol token's holder base is.
Q: What's Tron's TVL mostly made up of?
Tron's TVL is mainly JustLend (USDT lending) and SunSwap (a DEX). Underneath it all is a huge amount of circulating USDT — roughly $67 billion USDT sits on Tron (about 60% of total USDT supply), of which roughly $10 billion has flowed into DeFi protocols.
Q: Do Ethereum's L2s (Arbitrum, Optimism, Base) combined still add up to less than Ethereum mainnet?
Yes. Combined L2 TVL is roughly $13 billion, far below Ethereum mainnet's $72 billion. The main reasons: large capital, institutions, and LSTs are still deployed mostly on mainnet; L2s are dominated by retail and high-frequency trading; and some L2 assets (like USDC) are bridged over from mainnet, meaning they were already counted once there.
Q: How does DefiLlama handle private or closed chains?
It only tracks chains with public RPC nodes and readable contracts. Fully private chains, or systems that don't expose protocol data publicly, aren't included. That transparency requirement is the foundation of DefiLlama's data credibility.
Q: Will TVL keep growing going forward?
The medium-to-long-term outlook is positive, driven by a few catalysts: RWA growth (projected to top $50 billion by 2027), ETF inflows into crypto, growing institutional DeFi adoption, and falling interest rates making stablecoin DeFi yields relatively more attractive. But short-term swings remain large and track the broader crypto market.
The DefiLlama official dashboard, L2Beat, Token Terminal, Etherscan, Solscan, Tronscan, multiple Dune Analytics dashboards, and CoinGecko.