Kumo Research · Funding rates
Hyperliquid vs Binance funding rates: longs pay about twice as much on-chain (2023–2026)
Hyperliquid is one of the biggest on-chain perp exchanges, and many traders use it next to Binance. Funding works the same way on both, but it is set by different crowds. Hyperliquid pays every hour, Binance every 8 hours. We lined up both exchanges' funding day by day for 20 coins from June 2023 to October 2026, about 24,000 coin-days, to see where longs pay more, by how much and how often.
By Alex Wu · Published · 5 min read
Key findings
- Longs paid more on Hyperliquid on most days: the median coin had higher Hyperliquid funding on 71% of days. Bitcoin longs paid an average 14.3% a year on Hyperliquid against 7.2% on Binance, about twice as much.
- Across the 20 coins the average gap was 6.2 percentage points a year. It was biggest on meme and smaller coins: Pepe 13.3, WIF 12.9 and NEAR 13.0 points.
- The gap is not steady. On a typical day the two exchanges differed by about 7 points a year in one direction or the other, and on 17% of days by more than 20 points.
- Both exchanges have cooled. Pooled across coins, Hyperliquid longs paid 24% a year in 2024 and 4% in 2026; Binance went from 12% to 0.8%. The gap has shrunk in points but in 2026 Hyperliquid still pays about five times Binance's rate.
- A few coins are the exception. On ADA the two were even, and BNB and TRX had negative average funding on Binance while Hyperliquid stayed slightly positive. Where you trade changes what you pay.
Same idea, different crowds
Funding on Hyperliquid and Binance does the same job: it keeps a perpetual future close to the real price. When a perp trades above the spot price, longs pay shorts. Both exchanges also start from the same base of about 0.01% per 8 hours (roughly 11% a year). The differences:
- Hyperliquid pays every hour, Binance every 8 hours (every 4 on some coins).
- Different traders. Hyperliquid is on-chain, with no account sign-up, and is popular with crypto-native traders and bots. Binance has the largest mix of retail, funds and market makers.
- Different short supply. On Binance, big desks short the perp against coins they hold to earn funding (cash-and-carry). That is harder to do at scale on-chain.
So the question is simple: does the same coin cost more to hold long on one exchange than the other?
Bitcoin: about twice as much on Hyperliquid
Bitcoin funding: Hyperliquid vs Binance, month by month
Average funding paid by longs, % per year
- Hyperliquid
- Binance
Source: Hyperliquid API funding history; Binance public data archive · Chart: Kumo Research, kumocast.io
From June 2023 to October 2026, Bitcoin longs paid an average 14.3% a year on Hyperliquid and 7.2% on Binance. Hyperliquid was higher on 80% of days. Ethereum looks almost the same: 14.4% against 7.2%, higher on 76% of days.
The two lines move together. Both jumped in the early-2024 rally and both cooled in 2025–2026. But Hyperliquid's line sits above Binance's most of the time, and in calm months the gap is wide compared with the rate itself. In October 2026 Bitcoin longs paid 8.8% a year on Hyperliquid against 4.8% on Binance.
The gap, coin by coin
How much more longs paid on Hyperliquid, by coin
Average daily gap, Hyperliquid minus Binance, percentage points per year
- Pepe13.3
- NEAR13.0
- WIF12.9
- DOGE9.4
- AAVE9.3
- SOL7.6
- LINK7.5
- BTC7.1
- ETH7.1
- UNI5.9
- BNB5.1
- LTC4.7
- AVAX3.7
- DOT3.6
- ARB3.3
- SUI2.6
- TRX2.5
- XRP2.2
- BCH1.9
- ADA1.0
Source: Hyperliquid API; Binance public data archive · Chart: Kumo Research, kumocast.io
All 20 coins had a higher average on Hyperliquid, but by very different amounts. The biggest gaps were on the most speculative coins: Pepe (13.3 points a year), NEAR (13.0) and WIF (12.9). On those, eager longs crowd onto Hyperliquid while Binance has more shorts to balance them. At the other end, ADA was almost even (1.0 points) and Hyperliquid was higher on fewer than half of ADA's days.
| Coin | Days compared | Hyperliquid, % a year | Binance, % a year | Gap (points) | Hyperliquid higher | Gap over 20 points |
|---|---|---|---|---|---|---|
| Pepe | 1242 | +22.3% | +9.0% | +13.3 | 74% | 30% |
| NEAR | 1095 | +20.6% | +7.7% | +13.0 | 82% | 18% |
| WIF | 1017 | +23.7% | +10.7% | +12.9 | 81% | 22% |
| DOGE | 1242 | +17.2% | +7.8% | +9.4 | 70% | 22% |
| AAVE | 1202 | +16.3% | +7.0% | +9.3 | 83% | 20% |
| SOL | 1242 | +12.5% | +4.8% | +7.6 | 71% | 18% |
| LINK | 1242 | +15.8% | +8.3% | +7.5 | 76% | 20% |
| BTC | 1242 | +14.3% | +7.2% | +7.1 | 80% | 12% |
| ETH | 1242 | +14.4% | +7.2% | +7.1 | 76% | 13% |
| UNI | 1177 | +14.5% | +8.6% | +5.9 | 74% | 18% |
| BNB | 1242 | +3.6% | -1.5% | +5.1 | 73% | 16% |
| LTC | 1242 | +13.0% | +8.3% | +4.7 | 71% | 15% |
| AVAX | 1242 | +8.3% | +4.5% | +3.7 | 64% | 22% |
| DOT | 1145 | +7.0% | +3.4% | +3.6 | 58% | 14% |
| ARB | 1242 | +11.7% | +8.4% | +3.3 | 59% | 16% |
| SUI | 1242 | +8.3% | +5.8% | +2.6 | 61% | 20% |
| TRX | 1182 | +1.1% | -1.4% | +2.5 | 64% | 15% |
| XRP | 1232 | +9.2% | +7.0% | +2.2 | 62% | 14% |
| BCH | 1220 | -1.0% | -2.9% | +1.9 | 64% | 15% |
| ADA | 1105 | +8.5% | +7.4% | +1.0 | 46% | 13% |
Daily averages of each exchange's funding, converted to % per year (Hyperliquid hourly rate × 24 × 365; Binance 8-hour rate × 3 × 365). Pepe compares Hyperliquid kPEPE with Binance 1000PEPE (rates do not depend on the contract size).
It is not a steady gap
Averages hide how much the gap jumps around. On a typical day the two exchanges differed by about 7 points a year, and on 17% of days by more than 20 points. For Pepe that was 30% of days. Some days Binance is the expensive one.
Over the years both exchanges have cooled together:
- 2024: Hyperliquid 24.3% a year, Binance 12.3% (pooled across coins)
- 2025: Hyperliquid 7.9%, Binance 2.8%
- 2026 to October: Hyperliquid 4.2%, Binance 0.8%
The gap in points has shrunk, but as a multiple Hyperliquid has become relatively more expensive.
What this means
- If you hold longs for weeks, the exchange matters. At the average gap, a $10,000 Bitcoin long held for a year cost about $710 more in funding on Hyperliquid. On a meme coin it could be more than $1330.
- If you hold shorts, Hyperliquid usually paid you more. The same gap works in a short's favour.
- The gap is what funding arbitrage tries to collect (short where funding is high, long where it is low). It is real but not free: two sets of fees, margin on two exchanges, liquidation risk on either side and a gap that can flip. Our funding rate arbitrage page explains each risk.
Today's rates on Hyperliquid, Binance and 10 other exchanges are on the funding rates page, updated every 15 minutes. Our funding history study shows how Binance funding changed from 2020 to 2026.
Method, data and limits
Data: hourly funding history from Hyperliquid's public API and every funding payment for the matching USDT perps from Binance's public data archive, for 20 coins: BTC, ETH, SOL, XRP, DOGE, BNB, ADA, TRX, LINK, AVAX, LTC, BCH, DOT, SUI, NEAR, UNI, AAVE, ARB, WIF and PEPE. Comparison starts June 2023 (later for coins listed later) and runs to October 2026.
Method: for each coin and each UTC day where both exchanges have data (at least 20 hourly payments on Hyperliquid and 3 on Binance), we averaged that day's funding on each exchange, turned it into % per year and took the difference.
Limits: the gap is not free money. To earn it you would hold a short on one exchange and a long on the other. That means paying trading fees on both, keeping margin on both, and risking liquidation if the price moves fast on one side. The gap can also flip for days at a time. Hyperliquid's on-chain markets can also have different prices from Binance during fast moves.
Download the data (CSV, CC BY 4.0)
How to cite
Kumo Research (2026). "Hyperliquid vs Binance funding rates: longs pay about twice as much on-chain (2023–2026)". Kumocast. https://www.kumocast.io/research/hyperliquid-vs-binance-funding-rates
This study describes past market data. It is not investment advice and past behaviour does not guarantee future results.
Live data for these markets
More from Kumo Research
- BitcoinBitcoin on Fed, inflation and jobs-report days: how much the hour after the news really moves (2017–2026)6 min read →
- Funding ratesCrypto funding rates 2020–2026: longs used to pay over 35% a year, now they pay almost nothing6 min read →
- BitcoinBitcoin halving cycles compared: price, peaks and crashes after all four halvings (2012–2026)6 min read →
About the researcher
Alex Wu
Crypto, DeFi, perps and FX · driven by data
Alex designed the Kumocast model behind these signals, sets its rules, reviews its accuracy record and updates the methodology whenever the model changes. Signals are refreshed every 30 minutes with live market data.