Weekly Data Report Week of Sep 21 – Sep 27, 2026 API-verified ~15 min read

Hyperliquid Weekly Data Report — Week of September 21–27, 2026

Every metric below was fetched from a public API — the Hyperliquid official endpoint, DefiLlama, published venue tallies — on September 28, 2026, and the fetch is archived with a timestamp. This report prints only what a public source can verify: where a commonly cited metric has no public API, the table says so rather than estimating, and where a widely quoted figure uses a different construction, both are printed and named. The week in one line: HYPE printed a record $98.00 and closed $91.72, −2.2%; platform open interest reached $17.01B; and protocol fees for the full seven days totalled $19.46M (Sep 21–27). The seventh print, Sunday Sep 27, published at $1.75M shortly after this report’s Sep 28 fetch, lifting the six-day $17.71M that was current at fetch time to the full-week $19.46M.

The verified metrics

#MetricReadingSourceNote
1HYPE price (Sep 27 close)$91.72Hyperliquid candles−2.2% WoW ($93.75 → $91.72)
2Weekly high / low$98.00 / $89.77Hyperliquid candlesHigh printed Wed Sep 23, low Thu Sep 24
3All-time high (intraday)$98.00Hyperliquid candlesSep 23; the close sits −6.4% below it
4Market cap (at close)$20.4BDerived (222.446M × $91.72)222.446M circulating × Sunday close
5Circulating supply222.4MCoinGecko / venue talliesOf 955.3M total; 1B max supply
6Token volume (24h)$0.52BCoinGeckoSep 27 print; HYPE spot + perp token volume
7Perp OI — platform total$17.01BHyperliquid APICore $13.08B + builder markets $3.93B
8Perp OI — core order book$13.08BHyperliquid API178 live markets of 234 listed
9Perp OI — HYPE book$1.86BHyperliquid API20.59M HYPE × mark $90.31; −8.0% WoW
10Perp OI — HIP-3 builders$3.93BHyperliquid API23.1% of platform OI (was 23.5%)
11Live perp markets178 + 150Hyperliquid APICore + builder markets with seeded books
12Perp volume (24h)$5.15BHyperliquid APICore $4.31B + builders $0.84B
13HYPE funding (8h)~0.000%Hyperliquid API−0.0002%/8h; cooled from 0.039% — longs no longer pay
14Protocol fees (7d)$19.46MDefiLlamaSep 21–27; Sep 27 ($1.75M) published post-fetch — see ledger
15Protocol fees (30d)$69.3MDefiLlamaSeven-day run-rate annualizes to ~$1.01B
16Implied buyback capacity (7d)~$19.3MDerived~99% of revenue to AF (third-party estimate) × fees
17TVL — Hyperliquid L1$6.99BDefiLlamaUp 7.4% from $6.50B on Sep 20
18TVL — Arbitrum bridge$0.56BDefiLlamaDown 6.7% from $0.60B on Sep 20
19Total TVL (protocol)$7.54BDefiLlamaL1 + bridge; +6.2% WoW
20HYPE burned (cumulative)~48.7MPress tally~4.9% of max supply; no single-API ledger
212026 YTD revenue~$429MPress tallyFirst among crypto projects; narrower than gross fees (row 15)
22Next scheduled unlock14,175,778 HYPEPublished schedulesSep 29; ~1.4% of total supply, ~2.7% of market cap

Not verifiable via public API (printed as unavailable, not estimated): staked HYPE and staking APY, HLP vault TVL and return, Assistance Fund weekly USD purchases and cumulative ledger, weekly liquidations, 7-day perp volume, spot 7-day volume, active addresses, transaction count, HIP-3 weekly volume share as a single published series. Rows 20 and 21 are third-party tallies, labeled as such: no public endpoint publishes the burn ledger or a year-to-date revenue series, and they are not interchangeable with row 15.

The five-week chain

WeekCloseHighLowFees (7d)TVL L1 (close)
Aug 24–30$79.99$86.73$76.59$21.2M$6.43B
Aug 31–Sep 6$87.94$89.60$79.52$16.5M$6.41B
Sep 7–13$77.26$88.31$76.59$12.9M$6.33B
Sep 14–20$93.75$94.60$75.15$19.34M$6.50B
Sep 21–27$91.72$98.00$89.77$19.46M$6.99B

Seven-day total; Sep 27 ($1.75M) published post-fetch. W5 revised to $19.34M by the data provider after this report’s W5 publication.

The shape: a late-August breakout, two weeks consolidating under the $89.60 record, one −12.1% down week, a +21.3% reclaim, and now a place-holding week that printed the highest high and the highest low of the entire five-week window. The $89.77 low is the important structural change — it is 19.6% above the $75.15 low from the week before, which means the week that followed the record did not come close to retesting the level that the rally had launched from. The five-week fee total is ~$87.6M across the window, and the five-week price range is $75.15–$98.00.

Inside the week: the daily path

DateCloseΔProtocol fees (day)
Mon Sep 21$94.14+0.4%$4.65M
Tue Sep 22$97.27+3.3%$3.08M
Wed Sep 23$94.09−3.3%$2.75M
Thu Sep 24$92.08−2.1%$3.59M
Fri Sep 25$92.35+0.3%$2.47M
Sat Sep 26$92.59+0.3%$1.17M
Sun Sep 27$91.72−0.9%$1.75M

Two structural notes on the daily pattern. First, the week's shape is a single up-leg followed by five distributions: Tue Sep 22's +3.3% is the only session that added meaningfully, and the record high four sessions later was never matched by a close near it. Second, fee prints decoupled from price direction at the top — Monday's $4.65M was the largest single-day print since early August, and it printed on a +0.4% day, while Tuesday's +3.3% session produced only $3.08M. The fee line responds to how much a move is traded, not to how large it is, and Monday's print looks like the fee cost of positioning into a record rather than the revenue of one.

The same seven days for BTC and ETH

DateHYPE ΔBTC closeBTC ΔETH closeETH Δ
Sep 21+0.4%$86,662+6.7%$2,777+5.0%
Sep 22+3.3%$86,237−0.5%$2,756−0.8%
Sep 23−3.3%$84,366−2.2%$2,685−2.6%
Sep 24−2.1%$84,385+0.0%$2,687+0.1%
Sep 25+0.3%$84,066−0.4%$2,691+0.2%
Sep 26+0.3%$84,413+0.4%$2,697+0.2%
Sep 27−0.9%$84,437+0.0%$2,688−0.3%

Read across, and this is the week's most uncomfortable fact for HYPE holders: on Monday, the one session when the market ripped, HYPE did not participate. BTC added +6.7% and ETH +5.0% on Sep 21; HYPE added +0.4%. The token's own gains then came on the next day, in isolation, and were fully returned by Wednesday. Weekly totals on the Sunday-close clock: HYPE −2.2%, BTC +4.0% ($81,221 → $84,437), ETH +1.6% ($2,646 → $2,688). The week's high and low say the same thing in the other direction — BTC's $87,471 high and $80,888 low bracket a steady climb, while HYPE's $98.00 high and $89.77 low bracket a round trip.

The HIP-3 builder census, measured

Builder-deployed markets are measured directly from the API's perpDexs and metaAndAssetCtxs endpoints, summed across every registered dex. This week's census, run Sep 28:

Builder dexLive marketsOpen interest24h volume
xyz109$3.85B$0.80B
io8$0.06B$0.03B
para29$0.02B$0.01B
mkts4$0.01B$0.01B
Total (4 of 10 registered)150$3.93B$0.84B

The largest builder books this week: SP500 ($352.4M), SKHX ($330.7M), GOLD ($275.5M), SILVER ($237.8M), SKHY ($191.9M), XYZ100 ($186.2M), MU ($155.6M) and BRENTOIL ($148.0M) — a list that is almost entirely equities, metals and energy, i.e. the non-crypto side of the builder stack. Builder markets held 23.1% of platform open interest ($3.93B of $17.01B), down from 23.5% a week earlier — the second consecutive weekly decline in that share, both driven by the core book growing faster than the builder stack rather than by builder OI falling. Five of the ten registered builders have live markets; the other five show zero open interest and zero volume. On the total-OI question, see the ledger below — the platform figure here sums every live market, and widely published figures of $8.4–8.8B measure a different, narrower set.

Valuation arithmetic on verified inputs

From verified inputs only. Market cap at the weekly close ($91.72 × 222.446M circulating) is $20.40B. The seven-day fee total is $19.46M; annualised at 52 weeks that is a $1.01B run-rate, and the 30-day base ($69.3M × 52 ÷ 4.33) annualises to $0.83B. On fully diluted supply of 955.3M the same market value is $87.6B.

InputsValueMultiple on circulatingMultiple on FDV
Market cap / 7-day fee run-rate ($1.01B)$20.40B~20.2x~87x
Market cap / 30d fee base ($0.83B)$20.40B~24.6x~106x
Market cap / W5 7-day run-rate ($1.01B)$20.40B~20.3x~87x
Fully diluted value / 7-day fee run-rate ($1.01B)$87.6B~87xn/a

The multiple is now ~20.2x on the full seven-print basis — essentially unchanged from last week’s ~20.3x (W5, also seven prints). The 22.2x figure shown at fetch time was a missing-print artifact: it used only six of seven days on the numerator. The resolved 7-day multiple is marginally tighter than last week, not wider. Any single multiple quoted without its fee window is unfalsifiable; this table exists to make that point rather than to make a forecast. The FDV rows assume all 955.3M outstanding tokens trade at the same price, which is not how unlocks work.

Data quality ledger

Every open item in this report, stated plainly rather than smoothed over:

  • Sep 27 daily fee — resolved. DefiLlama published the final daily print ($1.75M) shortly after this report’s Sep 28 fetch. The full seven-day total is $19.46M, +0.6% versus W5’s $19.34M (both seven-print; W5 itself a small upward back-fill from the $19.25M published on Sep 21). The six-day $17.71M and the 22.2x multiple shown at fetch time were missing-print artifacts; the resolved 7-day multiple is ~20.2x, marginally tighter than W5.
  • Open interest — two constructions, both printed. This report's $17.01B sums every live market on the platform ($13.08B core + $3.93B builder) from the API. Third-party reporting during the week cited $8.4–8.8B, a figure that tracks closely with the venue's largest handful of books (its top three — BTC $3.14B, ETH $2.93B, HYPE $1.86B — already sum to $7.93B). Neither is an error; they are different measures, and quoting one under the other's label is how a $9B platform gets described as an $18B one.
  • Circulating supply — venue divergence. CoinGecko and CoinPaprika both put circulating supply at 222.4M, which is the basis for row 5's $20.4B market cap. Some exchange tickers list 251.2M and quote ~$23.4B. The difference is a definition of circulating supply, not of price, and it is why HYPE's market-cap rank varies by source this week.
  • All-time high — venue dispersion, much tighter than last week. The Hyperliquid candle series printed $98.00 on Sep 23; same-day coverage cited $97.96 and $97.95, and an intraday column on Sep 22 recorded $96.68 for a print that occurred earlier in the run. The spread is now roughly 1.4% between the highest and lowest record figures, against 4.0% a week ago, because the record was set in one liquid session rather than across two thin ones.
  • Resolved from last week: the HYPE perp open-interest and funding readings labelled "snapshot-time variance" in the previous report are this week replaced by single dated measurements, and the W5 fee figure now reads $19.34M on a full re-read against the $19.25M published on Sep 21 — a 0.5% back-fill by the data provider, disclosed rather than silently updated.

Frequently asked questions

Why is the fee figure for only six days?

DefiLlama published the Sep 27 daily print ($1.75M) after this report was fetched. At fetch time only six days ($17.71M) existed, and we published them without estimating the seventh. The resolved seven-day total is $19.46M, +0.6% versus W5’s $19.34M. The earlier “pending” framing was the as-fetched state, not an assumption.

Did open interest really rise while the token fell?

At the platform level, yes — $17.01B, up from $16.87B, because the core book grew from $12.90B to $13.08B. The HYPE book itself did the opposite, falling 8.0% to $1.86B as its price declined. That split is the point: the venue's derivatives stock is not one number, and the HYPE-specific slice is the one that tracks the token.

What does a funding rate of roughly zero mean?

That neither side is paying a meaningful premium to hold a position. A week ago longs paid 0.039% per 8h — a real cost that implied a crowded long book. At −0.0002% that cost has gone. It removes a source of forced selling and also removes the mechanical bid that a funding-positive regime creates; it is genuinely neutral rather than quietly bullish or bearish.

Why does this site never print a single "Hyperliquid open interest" without qualification?

Because the same three words describe at least two different sums that differ by roughly 2x, and the difference is entirely a definition. We print our construction, name the sources of the alternative, and let readers compare. A single unqualified number is the most common way to be both accurate and misleading.

Is this financial advice?

No. This is data analysis for informational purposes. Nothing here is a recommendation to buy, sell or hold HYPE or any other asset.

Related reading