Hyperliquid Weekly News Review — Week of September 28–October 4, 2026
This review covers the Hyperliquid ecosystem and the HYPE token for the week of September 28–October 4, 2026, in twelve stories. Every number below is pulled from a named primary source — the Hyperliquid API, DefiLlama, on-chain schedules, published framework specifications — and each entry names where it can be re-verified. The week in one line: HYPE absorbed the cycle's largest unlock — 14.18 million HYPE, roughly $1.2B — closed the week at $90.57, down 1.2%, and the protocol's second buyback engine delivered its first measured payment.
1. The week in twelve stories
Twelve items, each with the verified number first and an editorial take second. The takes are interpretation only — no take introduces a new figure; every fact above it is re-verifiable at the named source.
1. Monday's answer to a Tuesday question Sep 28
HYPE fell 4.6% on Monday Sep 28, from $91.72 to $87.47 — the day before the unlock. Coverage the same week (Lookonchain data, via Cryptopotato) documented a whale deposit of roughly $16M of HYPE to OKX and Bybit earlier in the run, and CoinMarketCap's daily analysis attributed the session to a combination of unlock positioning and profit-taking after the $98.00 record of the prior week.
Our take: the selling arrived one session early, which is the market's usual way of saying the event was priced. The interesting number is not the 4.6% — it is that Monday's drop took the token to $87.47, still 11.7% above the $75.15 low that the September rally had launched from. De-risking into a known supply event, from a level this far above the trend origin, is what absorption looks like before the fact.
2. The unlock lands Sep 29
On Tuesday Sep 29, 14,175,778 HYPE vested — approximately $1.2B at the week's prices, ~1.5% of total supply and ~2.7% of market capitalisation, split 46.6% insiders, 46.3% community and 7.0% foundation (published vesting schedules; Bitrue and Tokenomics.com tallies). HYPE closed the day at $86.03, down 1.6%, with an intraday low of $84.76.
Our take: the schedule's own historical average for the seven days after prior unlocks is −8.3%. The actual day-one answer was −1.6%. The distinction that matters is the one the coverage kept making and markets kept ignoring: 14.18M tokens becoming eligible is not 14.18M tokens becoming sold. The recipients' behaviour, not the schedule, is the event.
3. The absorption day Sep 30
Wednesday Sep 30 printed the week's lowest intraday price — $84.55 — and then reversed to close at $90.94, up 5.7%, the week's largest single-session gain (Hyperliquid candles). The token never traded below $86 again.
Our take: this is the session the unlock narrative gets graded on. A supply event resolves not on the day the tokens vest but on the day the market shows who is standing under the book. $84.55 was probed once, held for less than a session, and the reclaim added 7.5% off the low by the close. Whatever insiders did with their allocation, enough other demand existed to price it.
4. HYPE goes multichain Oct 2
On Friday Oct 2, HYPE was deployed on Solana, Base and Unichain via Wormhole's Native Token Transfer (NTT) framework, enabling native holding and trading of the asset across the three ecosystems (TradingView and CoinMarketCap coverage of the deployment).
Our take: accessibility is a demand-side programme. A token that could only be held in one place now exists wherever its potential holders already are. The measured effect on flows will take weeks to show up anywhere we can verify — but the direction of the change is unambiguous: the friction between HYPE and three large user bases dropped to zero on Friday.
5. The week's largest fee print came on a quiet day Oct 2
Friday Oct 2 also produced the week's largest protocol fee print: $4.06M, on a session where HYPE moved just +1.6% (DefiLlama dailyFees; Hyperliquid candles). The week's biggest price move — Wednesday's +5.7% — printed only $2.19M.
Our take: the fee line tracks how much a move is traded, not how large it is. Friday's print is the fee cost of a market repositioning after the unlock resolution — and, plausibly, the first full day of volume around the multichain deployment. It is a reminder that the revenue series and the price series measure different behaviours.
6. AQAv2 pays: the first measured distribution Oct 3
On Saturday Oct 3, the first AQAv2 reserve-yield distribution landed: approximately $14.58M in USDC routed to the Assistance Fund for HYPE buy-and-burn. The payment covers yield accrued from August 26 to September 24 on the platform's USDC reserves, at an implied average rate of about 3.14% — an annualised contribution of roughly $193M at current scale (Hyperdash co-founder Hans, via GetChain News; Crypto Briefing and The Defiant reporting on the framework).
Our take: this is the week's structural event, bigger than the unlock it shared the calendar with. Since we started this site, the reserve-yield line has been a modeled number — an estimate living in a table with a caveat attached. On Saturday it became a measured one, with a date, an amount and an implied rate. The estimate band we carried ($135M–$182M/yr) has been replaced by a first print that annualises above the top of it.
7. The combined buyback engine, quantified Oct 3
With the reserve-yield stream measured, third-party estimates now put the combined annualised buyback capacity above $900M: roughly $771M/yr from trading fees (with ~99% of order-book fees routed to the Assistance Fund) plus the reserve-yield contribution (CoinGecko News syndicated coverage, Oct 3; framework parameters per the AQAv2 specification).
Our take: treat the $900M as an extrapolation from a single 30-day print, not a run-rate — the first payment covered a window when reserves and rates were both known, and neither is guaranteed to repeat. But the composition change is real and permanent: the buyback no longer depends solely on trading volume. Margin that sits parked now buys HYPE too.
8. ETF flows: divergence, measured Oct 3
Weekly flow data for the window ending Oct 2 showed U.S. spot HYPE ETFs with a net inflow of $3.4M, against sharply reduced Bitcoin ETF inflows of $82.9M and Ethereum ETF net outflows of $118M (CoinMarketCap weekly flow coverage).
Our take: the absolute number is small and should not be dressed up — $3.4M against a $20B market cap is noise. The relative number is the story: in a week when the two largest ETF complexes bled or stalled, the newest wrapped asset in the set kept a positive sign. Institutional demand that persists through a supply event is worth more than institutional demand that appears only in up-weeks.
9. The treasury keeps buying Oct 3
Hyperliquid Strategies, the Nasdaq-listed HYPE treasury company, purchased an additional 1.9 million HYPE worth $167.2M (≈$88 per token), bringing its total holding to approximately 37 million HYPE (~$3.26B) plus $292.6M in cash (PANews, Oct 3, citing Hyperliquid News).
Our take: a listed vehicle adding $167M of a token in the same week the market absorbed a $1.2B unlock is the cleanest possible illustration of why the unlock's bear case under-specifies demand. The arithmetic of this market is a race between scheduled supply and accumulated structural buyers — treasuries, ETFs, the Assistance Fund — and this week both sides showed up with numbers.
10. The regulatory read Oct 3
Bitwise CIO Matt Hougan published analysis arguing that the Senate's failure to advance the CLARITY Act (Sep 15) is net-positive for tokens like HYPE, because the current path allows favourable guidance — including around buyback-based value accrual — to be implemented faster than the bill's stricter framework would have allowed (Crypto.news, carried by CoinMarketCap, Oct 3).
Our take: a fund manager whose firm sponsors a HYPE ETF arguing that less legislation is good for HYPE deserves the skepticism any interested analysis gets. But the underlying mechanism is worth tracking regardless of the source: Hyperliquid's token model routes revenue to holders in a way that sits awkwardly inside several proposed frameworks. The regulatory question has not gone away; it has just been deferred again.
11. Funding rebuilt from zero Sep 28 – Oct 4
HYPE perp funding, which had decayed to ~0.000%/8h at the end of September, rebuilt to +0.010%/8h by the week's end, while the HYPE perp open-interest book fell 5.5% to $1.76B (Hyperliquid API, fetched Oct 5).
Our take: the combination is more informative than either number. Positioning shrank through the unlock — longs closed or were flushed — and what remains is paying a small, positive premium again. A book that is smaller and paying funding is healthier than a book that is larger and free: the crowded long that defined the record run has been cleared out, at a cost of only 1.2% on the week.
12. The post-unlock scorecard Sep 29 – Oct 4
Measured from the unlock-day close ($86.03, Sep 29) to the Sunday close ($90.57, Oct 4), HYPE returned +5.3% in the days after the unlock — against the schedule's historical average of −8.3% over the seven days following prior unlocks (Tokenomics.com unlock history; our candle arithmetic).
Our take: one observation is not a distribution, and August's unlock was also absorbed (+3.0% on the day) before the tape rolled over in mid-September. But the pattern across the two largest unlocks of the cycle is now consistent: both landed into demand, and neither produced the drawdown the historical average promised. The base rate may be due for a revision — or the regime that produced it may have changed.
2. What the week changed
Three things are different at the Sunday close than at the Monday open. First, the supply question: the cycle's largest unlock is now a past event with a measured outcome — absorbed, with a +5.3% post-event return — rather than a forward risk with a −8.3% historical average attached. Second, the buyback question: AQAv2's first payment converts the reserve-yield line from modeled to measured, at an annualised pace (~$193M) above the top of the prior estimate band, and the combined buyback capacity now has two independent revenue legs. Third, the distribution question: HYPE now exists natively on three additional chains, which does nothing measurable this week and potentially everything measurable next quarter.
Against that, the tape itself was unimpressive: −1.2% on the week against BTC +2.4% and ETH +1.4% — a second consecutive week of underperformance against both majors — and weekly fees of $17.92M, down 22.5% from the prior week's revised total. The structural news was good; the flow news was not.
3. What did not break
The platform's operating metrics held through the supply event. Open interest across every live market stood at $16.69B at the week's end (core $12.80B + builder markets $3.90B), with the builder share actually rising to 23.4% from 23.1%. Total protocol TVL of $7.24B gave back 4.0% — worth watching, but modest next to the week's supply noise. The core order book's top three markets (BTC $3.30B, ETH $3.07B, HYPE $1.76B) carried $8.13B of the $12.80B core total. And the record high — $98.00, set Sep 23 — still stands, with the Sunday close 7.6% below it. Nothing in the week's price action came from platform distress; it came from a scheduled supply event and a fee series in its usual post-spike decay.
4. Sources & method
Prices, open interest, funding and volume: the Hyperliquid official API (candleSnapshot, perpDexs, metaAndAssetCtxs), fetched Oct 5, 2026, archive timestamped. Fees and TVL: DefiLlama (dailyFees, protocol endpoints). Unlock amount, recipient split and historical impact average: published vesting schedules as tallied by Tokenomics.com and Bitrue. AQAv2 payment: Hyperdash monitoring via GetChain News, with framework parameters from the AQAv2 specification as reported by Crypto Briefing and The Defiant. Treasury purchase: PANews citing Hyperliquid News. ETF flows: CoinMarketCap weekly flow coverage. Multichain deployment: TradingView and CoinMarketCap. Where a claim comes from press rather than an API, the story says so. DefiLlama revised the prior week's fee total upward after our Sep 28 publication ($19.46M → $23.13M); the week-over-week comparison in this review uses the revised figure, and the data report carries the full note.
Frequently asked questions
The unlock was $1.2B — why did HYPE only fall 1.2% on the week?
Because unlocked is not sold. The 14.18M tokens became eligible on Sep 29; what moved the price was whatever the recipients actually did, and the tape says the net flow was absorbable. Monday's −4.6% was positioning ahead of the event; the event day itself fell 1.6%; and the next session reversed 5.7% off the week's low. The total week-over-week move (−1.2%) is smaller than the pre-event move, which is the signature of a priced-in event resolving.
Is the $193M annualised AQAv2 figure reliable?
It is an extrapolation from one 30-day print ($14.58M covering Aug 26–Sep 24, implied rate ~3.14%). The inputs that will move the next payments — average USDC balances and short-term rates — are both variable, and the framework's own reserves estimate ($5B–$6.7B) is a range. Treat the first payment as a floor for the mechanism's existence, not a ceiling for its variance. The second payment, due around Nov 2, is the first data point on the distribution.
Why did weekly fees fall 22.5% if the platform is fine?
Two reasons stacked. The comparison base moved: DefiLlama revised the prior week's total upward to $23.13M after our publication, and like-for-like comparisons now use that figure. And volume decayed: the post-record week traded the positioning, this week traded the event, and fee prints follow traded volume, not headlines. A −22.5% fee week after a record week is decay, not distress — the 30-day total is $96.5M, the highest 30-day window in the platform's history outside August's peak.
Does multichain HYPE dilute the original token?
No. Wormhole's NTT framework mints representations against locked originals — the Solana, Base and Unichain balances are backed 1:1 by HYPE held in the framework's custody. Supply does not change; the locations where supply can be held and traded do. What changes is competition for liquidity across venues, which is a market-structure question rather than a tokenomics one.
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.