Kalshi's 7% APY on perps: the real rule, the math, and the trap we fell into

We put real money behind this before writing it. Since October 18 the lab has held a Bitcoin perp short on Kalshi, hedged with real Bitcoin bought on Coinbase. In its first two days Kalshi paid the short $0.22 in funding across five payments, and the position qualifies for the new 7% program. Small numbers, real money, every cent in Kalshi's own records.

Kalshi's app now says "Earning up to 7% APY" on perpetual futures. It sounds like 7% on your cash. It isn't, quite, and the difference decides whether you get 7% or something closer to 4%.

The rule, in one line

From Kalshi Prime's program terms, effective October 15, 2026:

Eligible Positions = min(open position size ÷ 1.5, your US dollar margin cash)

That amount earns 7% a year (counted on a 360 day year, so a touch more than 7%). Any cash above it earns the regular 3.5%. Interest is paid on cash, never on leverage: a $500 position backed by $100 earns on $100.

Three examples

So to earn the full 7% on all your cash, your position needs to be at least one and a half times your cash. With no position at all, cash earns the ordinary 3.5%.

The fine print that matters

• Which markets count: Bitcoin, Ethereum, Solana, Hyperliquid, gold, silver, US Large Cap and US Tech 100 perps.
• When it is checked: once a day, at a time Kalshi does not announce. A position opened and closed between checks earns nothing.
• Long and short in the same market cancel out. You cannot farm it by holding both sides on Kalshi.
• When it is paid: each month's amount arrives within 10 business days after the month ends.
• When it ends: the earliest of August 28, 2027, the moment Kalshi has paid out $7 million in total, or Kalshi ending it with one week's notice. Terms can change with one day's notice.
• No minimum balance for the 7% part. The $250 average only applies to the regular 3.5% on cash.

The trap we fell into

When we added to our short through the app's quick order screen, Kalshi opened it as a separate isolated position backed by only $40 of its own margin. Our main position could survive Bitcoin rising about 66%. The new one would have been force closed at a rise of about 13%. The order screen showed a leverage slider, but nothing said the order would become a separate position with its own, much closer, liquidation price. We closed it and added the contracts to the main position instead.

If you add to a perp position on Kalshi, check afterwards that you still have one position, not two, and look at each one's liquidation price.

Interest is the small part

The 7% on our cash is worth about $21 a year on this test. The bigger income is funding: on a perp, the side the crowd is crowding pays the other side every eight hours. On Kalshi lately, longs have paid shorts.

That is the trade we are running: short the Kalshi perp, hold the same amount of real Bitcoin, so price moves cancel and the funding plus the 7% are what is left. It is not free money. Funding swings, and in months like April 2026 it went negative, which is when a short pays instead of earns. And the 7% is a promotion with a budget and an end date.

What we will publish next

The first 7% payment should land by mid November. We will show the exact amount against the formula, along with every funding payment, in the weekly State of the Lab.

None of this is advice. It is a lab writing down what it found, with its own money on the line.