Ok so the quote that there are 2^96 private keys is wrong then

I realize this is an old discussion but this thread now appears towards the top of searches for "How many bitcoin addresses are there."

Here are some exact numbers for Bitcoin and Ethereum... and all of their relatives.

The total possible number of addresses is exactly

**2^160**.

As a decimal number (what most people consider "normal") this is 1,461,501,637,330,902,918,203,684,832,716,283,019,655,932,542,975.

As a hexidecimal, this number is: FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF

The total number of possible private keys is frequently listed as

**2^256** and for all sensible discussion, this is correct.

However, technically there are a few less because of the secp256k1 Curve usage. Words cannot express how insignificant of a difference this is from

**2^256**. Imagine all of earth's beaches. Now imagine them with one less grain of sand. Even that is overstating the difference.

But since we're talking exact numbers, here are the exact number of private keys possible. I'll list them compared to 2^256

Decimal

115,792,089,237,316,195,423,570,985,008,687,907,852,837,564,279,074,904,382,605,163,141,518,161,494,336 (exact private key maximum)

115,792,089,237,316,195,423,570,985,008,687,907,853,269,984,665,640,564,039,457,584,007,913,129,639,935 (2^256)

Hexidecimal

FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFE BAAE DCE6 AF48 A03B BFD2 5E8C D036 4140 (exact private key maximum)

FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF FFFF (2^256)

Going back to the discussion of how long it would take to brute force attack a specific address -- quite simply not possible with today's technology. The quote of 2^96 represented how many potential PRIVATE keys would work for a

* single PUBLIC key* if they were

*evenly distributed *(not how many total private keys were there). A brute force attack would require an approach of using all 2^256 since there's no way to tell if a private key will generate an address that's been used or not. At some point (2^200?) all 2^160 addresses would be accounted for, but again, impossible with today's technology.