How Much of Hyperliquid Trades Through Third-Party Apps?
Apps built on Hyperliquid sent 12.4% of its trading volume in October 2026, the highest full month on record. See the monthly share and the apps behind it.

You can trade on Hyperliquid without ever opening Hyperliquid. A wallet like Phantom or MetaMask, a phone app like fomo, a trading terminal like HyperDash: each one sends your order to the same order book, the shared list of buy and sell offers, and adds a fee of its own.
So how much of Hyperliquid's volume now arrives through third-party apps?
In October 2026 it was 12.4%: $31.1 billion of the $250.1 billion traded on Hyperliquid that month. That's the highest of the 15 full months on record, and the third month of growth in a row.
The short version:
• A record month. Apps sent $31.1 billion to Hyperliquid in October 2026, passing the earlier high of $30.5 billion from September 2025.
• Nearly double since July. App volume rose 93% from July to October. Hyperliquid's total volume rose 10% over the same months.
• Two apps drove nearly half the growth. HyperDash and fomo account for 48% of the increase.
• Without bots, apps' share is far higher. In an earlier study, automated traders were 71.9% of the volume placed without an app. If that still holds, apps were about a third of October's volume from everyone else.
The numbers come from Flowscan, the Hyperliquid explorer built by Hydromancer, and cover complete days up to 12 November 2026. The method and its limits are at the end.
What counts as a third-party app
Hyperliquid lets any app attach a builder code to the orders it sends. The code is an address that marks the order as coming from that app, and it lets the app charge a fee the user has approved, up to 0.1% on perps. (Perps are futures contracts with no expiry date. They're most of what trades on Hyperliquid.) Our builder codes guide explains how the approval and the fee work.
In this article, an app is any builder code, and app volume is the value of every fill on an order that carried one. A fill is one completed match between a buyer and a seller, and a large order can fill in several pieces. Orders placed on Hyperliquid's own site, or sent straight to its API by a trading bot or a market maker (a firm that keeps buy and sell offers on the book), carry no code. They make up the rest.
The share of Hyperliquid volume that comes through apps, by month
| Month | Through apps | All of Hyperliquid | Share |
|---|---|---|---|
| August 2025 | $22.1B | $422.8B | 5.2% |
| September 2025 | $30.5B | $295.9B | 10.3% |
| October 2025 | $28.3B | $335.6B | 8.4% |
| November 2025 | $25.9B | $262.2B | 9.9% |
| December 2025 | $18.7B | $178.1B | 10.5% |
| January 2026 | $16.7B | $218.0B | 7.6% |
| February 2026 | $16.6B | $211.6B | 7.8% |
| March 2026 | $19.2B | $219.0B | 8.8% |
| April 2026 | $15.4B | $190.7B | 8.1% |
| May 2026 | $16.0B | $205.2B | 7.8% |
| June 2026 | $20.0B | $272.9B | 7.3% |
| July 2026 | $16.2B | $227.3B | 7.1% |
| August 2026 | $21.1B | $248.0B | 8.5% |
| September 2026 | $25.8B | $255.0B | 10.1% |
| October 2026 | $31.1B | $250.1B | 12.4% |
| 1 to 12 November 2026 | $15.9B | $124.9B | 12.7% |
Three things stand out.
The share has moved between 5.2% and 12.4%. The low was August 2025, the first full month in the data and the busiest one for Hyperliquid as a whole. Since then no full month has been below 7%.
The share can rise in a quiet month. December 2025 reached 10.5%, yet apps sent only $18.7 billion that month. Total volume on Hyperliquid was $178.1 billion, its lowest of the period, so apps held a bigger slice of a smaller total.
October 2026 is a different kind of high. Hyperliquid's total stayed between $248 billion and $255 billion a month from August to October, and apps grew inside it: $21.1 billion, then $25.8 billion, then $31.1 billion. The first 12 days of November are running at 12.7%.
Which apps moved the number
These five apps added the most volume between July and October 2026:
| App | Type | July 2026 | October 2026 | Change |
|---|---|---|---|---|
| HyperDash | Trading terminal | $0.24B | $4.10B | +$3.86B |
| fomo | Social trading app | $0.45B | $3.74B | +$3.29B |
| Borsa | Trading terminal | $0.05B | $1.44B | +$1.38B |
| MetaMask | Wallet | $0.78B | $2.07B | +$1.29B |
| Rabby | Wallet | $0.90B | $1.93B | +$1.04B |
| All apps | $16.16B | $31.12B | +$14.96B |
The biggest gains came from apps that were small, or didn't exist, a year ago. HyperDash is a terminal for active traders. fomo is a social trading app for phones, and its first Hyperliquid trade was in June 2026. Together they account for 48% of the increase.
The wallets grew too. MetaMask and Rabby each traded over $1 billion more in October than in July, and Phantom, the largest app of all time, went from $2.09 billion to $2.71 billion.
Volume isn't the same ranking as fees. For the apps that earn the most, see our top builders by revenue.
Terminals bring the volume, wallets earn the fees
Here is the same app volume split by type of app, for the 30 days to 12 November 2026:
| Type of app | Share of app volume | Share of builder fees |
|---|---|---|
| Trading terminals | 33.3% | 11.1% |
| Wallets | 26.5% | 49.9% |
| Social trading apps | 13.2% | 15.5% |
| Mobile trading apps | 3.6% | 3.7% |
| Copy trading apps | 2.0% | 2.0% |
| Everything else, including unlabeled codes | 21.4% | 17.8% |
Trading terminals sent 33.3% of app volume and collected 11.1% of the builder fees. Wallets sent 26.5% and collected 49.9%. Terminals serve fewer, larger traders at low fees. Wallets serve many smaller traders and charge more per dollar traded. In those 30 days, apps collected $13.3 million in builder fees between them.
If you're weighing what to build, we have guides to a trading terminal, a Telegram trading bot and a copy trading bot on Hyperliquid.
How concentrated app volume is
More than 2,100 builder codes have sent at least one trade to Hyperliquid, and 1,022 of them did in the last 30 days. Volume is much more concentrated than that:
• The 5 largest apps sent 49.9% of app volume in those 30 days.
• The 10 largest sent 69.5%.
• 221 codes sent more than $1 million, 37 sent more than $100 million and 9 sent more than $1 billion.
Why 12.4% understates what apps do
"All of Hyperliquid" includes market makers and trading bots, which trade far more than people do and mostly don't go through an app.
In our study of how profitable Hyperliquid frontend traders are, we filtered Hyperliquid's own vaults and automated traders out of every perp trade from August 2025 to April 2026. They were 0.48% of users. They were also 71.9% of the volume traded without an app, and 5.6% of the volume traded through apps.
Apply those two rates to October 2026 and apps come to about 32% of the volume from accounts the study didn't flag as automated. That's an estimate, because the rates were measured on perps in earlier months. It still shows the gap: with automated trading set aside, apps are a much bigger door into Hyperliquid than 12.4% suggests.
What this means if you're building an app
• The door is getting wider. Apps nearly doubled their volume in three months while the exchange's total rose 10%.
• New apps still break in. fomo went from its first trade to the second-largest app by 30-day volume in five months.
• Most of the exchange still trades without an app. In October, 87.6% of volume carried no builder code.
Check the number yourself
Without code. Each app's page on Flowscan shows its share of Hyperliquid volume and its share of all app volume.
With code. Hydromancer's Builder Dashboard API returns both totals alongside any app's own stats. Send your API key in the Authorization header, as the example does. The builder address below is Phantom's, and any builder code works:
import os, requests
URL = "https://api.hydromancer.xyz/dashboard/builder-stats"
KEY = os.environ["HYDROMANCER_API_KEY"]
BUILDER = "0xb84168cf3be63c6b8dad05ff5d755e97432ff80b" # any builder code
r = requests.get(URL, params={"builder": BUILDER, "window": "30d"},
headers={"Authorization": f"Bearer {KEY}"})
r.raise_for_status() # stops here with the reason if the key is refused
s = r.json()
apps, total = float(s["allBuildersVolume"]), float(s["hyperliquidVolume"])
print(s["startDate"], "to", s["endDate"])
print(f"Through builder codes: ${apps / 1e9:.2f}B")
print(f"All of Hyperliquid: ${total / 1e9:.2f}B")
print(f"Share: {apps / total:.2%}")2026-10-14 to 2026-11-12
Through builder codes: $36.34B
All of Hyperliquid: $281.25B
Share: 12.92%The 30-day figure is a little higher than October's because it includes the first 12 days of November. Dashboard endpoints are switched on per API key, so ask the team if yours returns a permission error.
Method and limits
• Source. Flowscan and Hydromancer's Builder Dashboard API, which read the same data. We took the snapshot on 13 November 2026.
• App volume. Price times size, summed over every fill on an order that carried a builder code.
• Hyperliquid volume. Every trade on perps, markets deployed by outside teams (HIP-3), spot (buying the token itself) and outcome (prediction) markets, counting one side per trade.
• How the months were built. Each app's daily record includes its share of Hyperliquid volume and its share of all app volume that day. We worked back to the two totals from four large apps separately. The four agreed to within 0.00002% on every day, and the sum since 27 July 2025 matches the API's own all-time total: $341.9 billion of $3.97 trillion, or 8.6%.
• Overlap. A trade between two app users puts both fills in app volume and counts once in the total, which nudges the share up. We checked two apps: one week of a large app's fills, and every Pump.fun fill. Orders that took liquidity, meaning they traded against an offer already waiting on the book, were 96% of volume at the first and 99.5% at the second, and two such orders can't trade with each other. At the first app, trades with its own users on both sides were 0.015% of its volume. We haven't measured the overlap across every app.
• Dates. The fill history starts on 27 July 2025. July 2025 is left out, and November 2026 is a part month.
• Labels. App names and types are Flowscan's. "Everything else" includes 967 codes with no app type on Flowscan that traded in the 30 days.
• Fees. Builder fees paid in the dollar stablecoins USDC, USDT0 and USDE. That isn't a company's total revenue.
FAQ
How much of Hyperliquid's volume goes through third-party apps?
12.4% in October 2026, the highest full month so far. Since 27 July 2025 the figure is 8.6%: $341.9 billion of $3.97 trillion.
Which app sends the most volume to Hyperliquid?
In the 30 days to 12 November 2026 it was HyperDash, with 16.9% of app volume, followed by fomo with 12.7% and Phantom with 8.4%. Phantom has sent the most of all time.
Does trading through an app cost more than trading on Hyperliquid directly?
Yes. The app's builder fee is added to Hyperliquid's own fee, and it can be up to 0.1% on perps. Our Hyperliquid fees guide has the rates.
How many apps are built on Hyperliquid?
More than 2,100 builder codes have sent at least one trade, and 1,022 did in the 30 days to 12 November 2026.
Sources
#1: Flowscan, Hyperliquid builders
#2: Hydromancer docs, Builder Dashboard
#3: Hyperliquid docs, Builder codes
#4: Hydromancer, How Profitable are Hyperliquid Frontend Traders?
Read next

Pump.fun on Hyperliquid: The First Four Weeks in Numbers
Pump.fun perps run on Hyperliquid. In four weeks: $457.6 million traded by 10,008 wallets, with Nvidia stock the second most traded market.

Hyperliquid API Latency by Region
We streamed Hyperliquid's BTC order book from AWS and Vultr servers around the world, from both the native API and Hydromancer. How fast? The data is here.

Top HIP-3 Market Deployers by the Numbers
Compare eight HIP-3 deployers by trading volume, market focus, and current status as of Nov 2026.