Every claim about DefiLlama below was read on 2026-09-20 from its own documentation at docs.llama.fi and from its public API, not from a landing page. Where the docs do not say something, this piece says the docs do not say it, rather than claiming the feature is missing.
Does DefiLlama already cover crypto derivatives data?
Yes, and more thoroughly than most people assume. Its exchanges dashboard tracks spot, USD-margined perpetuals and coin-margined perpetuals per venue, publishes open interest for linear and inverse books separately, and carries a current perpetual funding rate normalized to a common 8-hour basis. What it is assembled to answer is still a different question from what a market-wide derivatives view answers, and that difference decides which one you should open.
The subject of the sentence is different
Both tools read the same raw inputs: venue volume, open interest, funding, order book depth. The difference is what each number is an attribute of.
On DefiLlama's exchanges dashboard the subject is the exchange. Volume, open interest, book quality and reserves are properties of a venue, and the output is a ranking of venues. On a cross-venue derivatives view the subject is the market. Funding and positioning are properties of an instrument, the venue is one dimension of the grid, and the output is a reading of where crowding sits right now.
That is why they are not substitutes. A ranking of venues cannot carry a claim about an instrument, and a grid of instruments cannot carry a claim about a venue's solvency. The row you are allowed to cite is the row whose unit matches the sentence you are writing.
What DefiLlama's exchanges dashboard is built to answer
Its methodology page states two principles up front, verbatim: "We store what the exchange said, verbatim. Nothing is corrected on the way in." The sentence after it explains why, that every adjustment happens at read time from a declared rule "so the original record stays auditable and a rule can be revised without rebuilding history."
The second principle is one this industry gets wrong constantly: "A missing measurement is never scored as a bad one. Where something has not been measured it reads as unavailable, not as zero."
On units, the page says segments "are never merged into a single headline without being named, because they are different instruments with different notional conventions", and it applies a per-venue double-count correction where a venue reports both sides of a position as open interest. That is a stricter public disclosure than most paid vendors publish, and it deserves saying plainly.
One scope note, stated carefully because the careless version would be wrong. DefiLlama does publish liquidations: defillama.com/liquidations was live on 2026-09-20, titles itself a DeFi Liquidations Dashboard, and covers lending protocols such as Aave, Compound, Morpho and Venus. Read anonymously it answers "An active subscription is required to view liquidations data", and no public API endpoint for it answered. Those are on-chain lending liquidation levels. What the exchanges methodology does not describe is a liquidation number for the centralized venues it scores. Two different populations, which is the point of this piece rather than a mark against anyone.
What a market-wide derivatives view is built to answer
Read anonymously on 2026-09-20, the Athenum funding endpoint in per-exchange mode returns five named venues in that response, across thirty symbols, with the values differing per venue on the same symbol rather than collapsing into one aggregate column. That grid is the shape of the object: same instrument, several boards, one screen, so a difference between boards is visible instead of averaged away.
The free entry point is the calculator set, and our own tools page states it without hedging: "All 34 tools are completely free with no account creation, no email capture, and no usage limits." Several of those calculators do pull live prices. The terminal is a separate thing and it is worth being exact about: its free tier, listed as Agora on our own pricing cards, is delayed by fifteen minutes, and real-time streaming starts on the first paid tier. This piece will not pretend otherwise.
Side by side
| What is being measured | DefiLlama exchanges dashboard | Athenum |
|---|---|---|
| The unit one row describes | One exchange | One symbol on one venue, inside a grid |
| Reserves and solvency | A reserves pillar carrying 30 of 100 points in its REAL score | We do not measure exchange solvency |
| Execution quality | Effective spread, markouts and slippage on a shared basket | Not scored; resting book depth is read as market structure |
| Open interest | Per venue, linear and inverse published separately | Per symbol and venue, for cross-venue comparison |
| Funding | Per venue, normalized to a common 8-hour basis | Per symbol and venue, in one grid |
| Liquidations | A DeFi lending liquidations dashboard, subscriber gated | A model of centralized venue liquidations, built from public data |
| Unit handling | Declared per venue, market and contract type, with a double-count correction | Declared per series, with the same inverse and linear problem |
| Coverage breadth | 8,313 protocols and 467 chains from the public API on 2026-09-20 | Derivatives venues and symbols, not the DeFi long tail |
| Cost to start | Free dashboard, with Pro at 49 dollars a month and API access at 300 dollars a month | Free calculators with no account |
Where DefiLlama is the stronger tool, plainly
Three places, and none of them are close. Breadth: its public protocols endpoint returned 8,313 entries and its chains endpoint 467 on the day this was written, which no venue-polling terminal will ever match for DeFi. Solvency: the reserves pillar answers a question we do not answer at all. Openness: adapters are community contributed through a public repo, and the docs say so, which means coverage grows without anyone's sales team.
That last point has a flip side worth knowing rather than hiding. When coverage arrives by contribution, a gap can mean nobody wrote the adapter yet, not that nothing is happening. Read an absence there as unmeasured, which is precisely what the methodology page tells you to do.
What each one can actually support
A number is only worth citing if the instrument that produced it can carry the sentence you are writing. A claim that an exchange is liquid, that its reserves are covered, that its fills are cheap, is a claim about a venue, and the venue scorecard is what supports it. A claim that positioning has crowded onto one board, or that funding flipped on one symbol and not its neighbour, is a claim about an instrument, and no venue ranking can support it no matter how good the ranking is. The common failure is not picking the wrong tool, it is citing one of them for a claim only the other can carry.
One dating note, because this field is moving fast. DefiLlama shipped its REAL exchange ranking on 2026-09-16, four days before this was written. Treat every line above as a reading taken on 2026-09-20 and go and re-read it yourself rather than quoting this page back later.
If the second kind of claim is the one you need, Athenum's per-venue turnover breakdown is a worked example of reading one market across several boards.