Over 107 days of 20-level order-book snapshots, xyz:ZM quoted a median spread of 9.77 bps in US regular hours with $141k resting within 25 bps of mid. A $25k market order paid about 7.39 bps in regular hours and 35.41 bps on weekends, when depth within 25 bps fell to $9k. Taker fee is 0.90 bps with growth mode on; funding averaged +6.19% annualised over the last 30 days.
spread · rth
9.77 bps
median · p90 17.14 bps
depth ±25bps · rth
$141k
median, both sides
$25k slip · rth
7.39 bps
fills 97%
$25k slip · wknd
35.41 bps
fills 31%
taker fee
0.90 bps
fee scale 1.0 · growth mode on
funding apr · 30d
+6.19%
longs pay when positive · 7d +2.11%
vol 24h · 30d avg
$108k
notional
open interest
$461k
both sides · 2026-10-02
by sessionNew York time · medians over 107 days
session
snaps
spread
p90
top
depth ±10
depth ±25
depth ±50
$5k
$25k
$100k
$100k fills
$250k fills
US regular 9:30–16:00
4,899
9.77
17.14
$6k
$55k
$141k
$249k
5.79
7.39
16.54
62%
3%
US extended 4–9:30, 16–20
7,165
16.66
29.51
$1k
$1k
$19k
$51k
16.35
27.51
53.32
10%
0%
Overnight 20–4
4,941
16.59
29.90
$1k
$1k
$12k
$32k
19.23
32.15
94.30
6%
0%
Weekend / holiday
9,123
19.80
33.58
$1k
$208
$9k
$23k
22.18
35.41
107.83
1%
0%
bps unless stated · RTH 9:30–16:00 · extended 4:00–9:30 and 16:00–20:00 · overnight 20:00–4:00 on weekdays · weekend runs Friday 20:00 to Monday 4:00 and includes NYSE holidays · slip = mean of the buy and sell side of a market order walking the visible 20 levels · fills = share of snapshots the size fits inside those levels · "—" = not fillable in the median snapshot
trade flowtaker side · 24 h to 2026-10-02 00:41 UTC · 30 d = last 30 complete UTC days · all markets
alert melarge orders in ZM, $50k or more→ sign infree · by email · when the data is rebuilt
taker buy · 24h
$59k
30d $1.5M
taker sell · 24h
$65k
30d $1.6M
imbalance · 24h
-5.4%
of $124k · 30d -5.4% of $3.1M
trades · 24h
481
0 large orders · 30d 34,096
takers · 24h
70
30 makers · 84 addresses
takers · 30d
401
431 addresses incl. makers
top-10 share · 30d
44%
of $3.1M by 401 takers · 13 make half
concentration · 30d
0.027
Herfindahl over takers · 1 = one taker
daily net taker flow
USD per UTC day · taker buy − taker sell · up = net buying, down = net selling · 108 complete days · hover a day for its value · today (2026-10-02) is still open and not drawn: −$22k so far
by session · 30 d
session
taker buy
taker sell
net
imbalance
trades
share of notional
US regular 9:30–16:00
$729k
$922k
−$193k
-11.7%
21,191
53%
US extended 4–9:30, 16–20
$362k
$312k
+$51k
+7.5%
5,488
22%
Overnight 20–4
$168k
$198k
−$29k
-8.0%
3,874
12%
Weekend / holiday
$217k
$214k
+$3k
+0.6%
3,543
14%
sessions in New York time as in the liquidity table above · share = the session's taker notional over all four, $3.1M in 30 days
hour of week · taker notional
New York time · 30 days summed per hour · brighter = more volume · log scale between the 5th and 95th percentile · 168 of 168 hours with trades · hover a cell for USD and trades
large orders: fills grouped by transaction hash, taker and side, $50k or more per order; fills without a hash are not grouped · share = the address's taker notional over all $3.1M of 30-day taker notional (34,096 trades, 401 takers) · addresses are public on-chain data and link to the Hyperliquid explorer; an address can be a subaccount, vault, market maker or bot and is not labelled · taker notional is volume, not a change in open interest · rebuilt nightly and at 09:05 ET, not a live tape · raw: /api/xyz/flow/ZM.json · definitions in methodology
alert meZM funding above or below a level I set→ sign infree · by email · when the data is rebuilt
last settlement
+66.23%
+0.7561 bps/h · 10-01 21:00 ET
next · estimate
+18.58%
+0.2121 bps/h · as of 10-01 22:15 ET
apr · 24h
+16.35%
15 of 24 hours observed
apr · 7d
+2.11%
157 of 168 hours observed
apr · 30d
+6.19%
537 of 720 hours · stdev 56.83%
paid by a long · 30d
37.92 bps
of notional, over 537 observed hours
hours above neutral · 30d
51%
at 16% · below 34% · negative 32%
range · 30d
+264%
highest hour · lowest -174%
daily funding apr
mean of the observed hourly rates per UTC day, annualised · up = longs paid, down = shorts paid · dashed = neutral 5.48% · 110 complete days · hover a day · today (2026-10-02) is still open and not drawn: +70.62% over 2 h so far
hour of week · mean funding apr
New York time · whole archive · the hour a settlement covers · 168 of 168 hours with data · hover a cell for its mean and n
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Mon
Tue
Wed
Thu
Fri
Sat
Sun
shorts paylongs pay · full shade = ±34.72% APR or more · empty = no settlements
funding by session
session
apr · 30d
hours · 30d
above neutral · 30d
apr · all
hours · all
above neutral · all
US regular 9:30–16:00
+15.29%
89
82%
+10.77%
403
58%
US extended 4–9:30, 16–20
+23.49%
157
61%
+13.04%
686
50%
Overnight 20–4
-0.69%
104
39%
-2.37%
453
40%
Weekend / holiday
-8.85%
187
34%
-0.53%
841
41%
mean of the hourly rates settled in each session, annualised · 30d = the 720 hours ending at the last settlement, all = since 2026-06-14 · above neutral = share of those hours settled above 5.48% APR
largest settlements · 30 d
settled · ET
session
rate bps/h
APR
09-30 07:0009-30 11:00 UTC
US extended
+3.017
+264%
09-14 07:0009-14 11:00 UTC
US extended
+2.408
+211%
09-13 21:0009-14 01:00 UTC
weekend / holiday
+2.323
+203%
09-03 09:0009-03 13:00 UTC
US extended
+2.269
+199%
09-30 17:0009-30 21:00 UTC
US extended
-1.99
-174%
09-22 07:0009-22 11:00 UTC
US extended
+1.889
+165%
09-10 22:0009-11 02:00 UTC
overnight
+1.839
+161%
09-06 21:0009-07 01:00 UTC
weekend / holiday
+1.829
+160%
09-24 17:0009-24 21:00 UTC
US extended
+1.779
+156%
09-18 09:0009-18 13:00 UTC
US extended
-1.754
-154%
positive = longs pay shorts · the neutral rate is the deployer's carry (base interest 0.01% per 8 h × funding multiplier 0.5), not crowd positioning · every mean is over observed hours; collector gaps are not filled · rebuilt nightly and at 09:05 ET, not live · raw: /api/xyz/funding/ZM.json · definitions in methodology
slippage curve
median bps over mid vs order size · a line ends where the size stops filling in the median snapshot
hour of week
New York time · median per cell over 107 days · brighter = better · the breathing cell is now
weekly · spread
median bps, all sessions
weekly · depth ±25 bps
median USD both sides, all sessions
cost calculator
per round trip = 2 × median slippage + 2 × taker fee · funding uses the 30-day mean hourly rate, annualised, charged for a position held all day · slippage between tested sizes is interpolated on a log scale · ignores maker rebates, impact that persists after your fill, and fee tiers