All Bitcoin address with balance, update weekly.

[quantummachine]

I couldn't find a complete list of all bitcoin address with balance on the web, so I decide to put up one myself, and I plan to update the list weekly.

https://bitkeys.work

I hope it's useful for someone who do casual research on Bitcoin address, to add a "fun" flavor, random private keys are generated to the right, with the slim hope that someone is lucky enough to find a collision.

As of May 7 2019, there are 24,065,464 Bitcoin addresses with balance, the top address has 124K bitcoins in it.

The serious part is: I have a little doubt on the viability of RIPEMD160, by the curse of "birthday paradox", approaching 2^80 computation doesn't seem to be too remote. It's just not economical (profitable). At present, the whole network can do 2^66 hashes per second, it takes 2^14 seconds - less than 1/4 day, to reach 2^80 computation. And remember, that is 2 times SHA256 on a block. ASIC's that are computing one-time random SHA256 should cost less and do more.

Looking ahead the expected long life span of Bitcoin, the exponential increase of mining difficulty, the flat/constant difficulty of generating SHA256 private keys, we may see the equation changed to favor private key searching than mining.

I welcome anyone to correct my math, better not to look like an idiot.  

One more thing, to help those who do intensive private key searching, I can send you the whole list as long as I can, so PM me. My last parsing on April 30 2019 returned 23+ million addresses, as you can see from the website, the top address has 125,805 BTC, all the way down to many addresses with only 1 satoshi in them...

Right now the file is in CSV, format is:

base58-address, balance-in-satoshi

Thanks again, I look forward to your comments, and corrections.

[1714687135]

1714687135

[1714687135]

1714687135

[1714687135]

1714687135

[1714687135]

1714687135

[Vod]

Your columns  Compressed  and Uncomp'd

It would be nice if you could display full address when you hover.

And the font is a bit small.  Other than that, leaving you a couple merit for your obvious knowledge of bitcoin. 

[AB de Royse777]

I left 4 from my bag.
So, these private keys, how do you come up with the keys? I assume none of them are their own private key.

Edit:
I am trying to understand the  Compressed  and Uncomp'd column. Pardon me not to have much knowledge.

What does those addresses means in the  Compressed and  Uncomp'd column? How are they connected with the address with the balance?

Edit1:
How about to have a search field with a given address? They way people can search with a known address and can see it in the list. Just a thought.

[quantummachine]

This is an explanation of Bitcoin's compressed and uncompressed versions of address.

https://github.com/OmniLayer/omniwallet/wiki/Converting-between-Compressed-and-Uncompressed-Addresses-and-Private-Keys

The takeaway is: each private key can generate 2 versions of Bitcoin address, compressed and uncompressed. When you stare at a Bitcoin address, you CAN'T tell whether it is compressed or not, so you have to check the balance of both versions for a single private key.

OK, I will tweak the UI, font size, search box etc.

Thanks for the feedback so far...

[Patatas]

It would be nice if you could display full address when you hover.

Actually, if you're using a Chrome browser, you can see the full link what the address points to in the bottom left part of your chrome window. It shows the entire address. Not sure if other browsers have that feature.

@OP: That is the most pagination I've seen in my life. Are these pages/tables generated on the fly or you receive them from the network as soon as the website is opened?

[TryNinja]

@OP: That is the most pagination I've seen in my life. Are these pages/tables generated on the fly or you receive them from the network as soon as the website is opened?

Obviously on the fly.

What OP probably does is generate some private keys on the user's end through Javascript based on which page you are; if the address is not empty, sends the private key to his server. Where he can send the coins to his own wallet.

Basically, everybody that tries this is using his own hardware power to brute-force private keys for OP. Free work.

[pooya87]

The serious part is: I have a little doubt on the viability of RIPEMD160, by the curse of "birthday paradox", approaching 2^80 computation doesn't seem to be too remote. It's just not economical (profitable). At present, the whole network can do 2^66 hashes per second, it takes 2^14 seconds - less than 1/4 day, to reach 2^80 computation. And remember, that is 2 times SHA256 on a block. ASIC's that are computing one-time random SHA256 should cost less and do more.

finding a hash collision for RIPEMD160 is impossible because the chances of it are astronomically small and it  doesn't do you any good because you need to have a private key that produces a public key that is then hashed with SHA256 and then hashed with RIPEMD160 that produces the same hash! that is the combination of 3 impossible things!

2^80 is huge and what miners are doing with their 2^whatever hashes per second is not a hash collision find!

[quantummachine]

It would be nice if you could display full address when you hover.

Actually, if you're using a Chrome browser, you can see the full link what the address points to in the bottom left part of your chrome window. It shows the entire address. Not sure if other browsers have that feature.

@OP: That is the most pagination I've seen in my life. Are these pages/tables generated on the fly or you receive them from the network as soon as the website is opened?

I parsed the blockchain, extracted all ~24 million addresses with unspent transaction output (UTXO), and dump them into DB table, you divide 24 million by 50 (per page), you get this gigantic pagination. But they are actually all in a local DB server, not generated on the fly.

The random private keys and their compressed and uncompressed addresses, are generated on the fly, on each page load.

Of course you don't search private keys by flipping web pages, it'd be too slow, so this project is just for fun. For professional searchers/hackers, you run parallel programs that use all your CPU cores and GPU ALU's, better written in Pony language, or C/C++/Cuda, for a slim hope to find a collision. And yes I agree, 2^80 is a huge number. And even when you produce 2^80 private keys, the probability that there exist collision among the 2^80 keys, is 50%. But when you possess this monster computational power, 2^80, I see Bitcoin is on the brink of collapsing.

[tyz]

I couldn't find a complete list of all bitcoin address with balance on the web, so I decide to put up one myself, and I plan to update the list weekly.

https://bitkeys.work

-snip-

And you are sure that the randomly generated addresses are not already checked by you for possible balance before they are displayed on the page? 
Would at least make sense or you really are a benevolent person.

[Thirdspace]

The takeaway is: each private key can generate 2 versions of Bitcoin address, compressed and uncompressed. When you stare at a Bitcoin address, you CAN'T tell whether it is compressed or not, so you have to check the balance of both versions for a single private key.

afaik, each Private Key (Hex Format) can generate 4 valid bitcoin addresses in 3 address formats (P2PKH, P2SH, Bech32)
1 is derived from uncompressed public key, and 3 more from compressed public key

for example,
Private Key Hexadecimal Format: 0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF0123456789ABCDEF

Private Key WIF uncompressed: 5HpneLQNKrcznVCQpzodYwAmZ4AoHeyjuRf9iAHAa498rP5kuWb
Public Key uncompressed: 044646AE5047316B4230D0086C8ACEC687F00B1CD9D1DC634F6CB358AC0A9A8FFFFE77B4DD0A4BF B95851F3B7355C781DD60F8418FC8A65D14907AFF47C903A559
P2PKH (legacy) uncompressed : 1CLrrRUwXswyF2EVAtuXyqdk4qb8DSUHCX

Private Key WIF compressed: KwFvTne98E1t3mTNAr8pKx67eUzFJWdSNPqPSfxMEtrueW7PcQz
Public Key compressed: 034646AE5047316B4230D0086C8ACEC687F00B1CD9D1DC634F6CB358AC0A9A8FFF
P2PKH (legacy) compressed: 1M8Qk46ERsPrEtWLBRSET5NUH2Ck5wwREU
P2SH Segwit Address: 31khcn4aNFxMdvcVUCHj8Z1u1A7CqBqyAr
Bech32 Address: bc1qmnyn7x24xj6vraxeeq56dfkxa009tvhgqffstc

[pooya87]

~

you missed P2SH (commonly known as multi signature but in this case with 1 sig or 1of1).
also you could do the same with bech32 addresses meaning a multisig SegWit bech32 address with only 1 key.
i am not sure if you could do the same with P2WSH since i haven't seen any multisig that way but that could be another one too.

[quantummachine]

I couldn't find a complete list of all bitcoin address with balance on the web, so I decide to put up one myself, and I plan to update the list weekly.

https://bitkeys.work

-snip-

And you are sure that the randomly generated addresses are not already checked by you for possible balance before they are displayed on the page? 
Would at least make sense or you really are a benevolent person.

For the record, no. I don't check it sneakily. In fact, each address generated is automatically linked to blockchain.com, it's a bit cumbersome but you do have the means to verify the balance on a 3rd party website.

[quantummachine]

It would be nice if you could display full address when you hover.

Actually, if you're using a Chrome browser, you can see the full link what the address points to in the bottom left part of your chrome window. It shows the entire address. Not sure if other browsers have that feature.

@OP: That is the most pagination I've seen in my life. Are these pages/tables generated on the fly or you receive them from the network as soon as the website is opened?

If you are interested in any address, you can click it to blockchain.com page, there you can have a better view of the transaction details of any address, if transactions happened before.

I can build a relational DB and provide the same view as blockchain.com, but I just feel they already did a good job and there is no point of repeating it.

My primary goal is to provide the whole list of BTC address with balance, which I can't find elsewhere.

[quantummachine]

The serious part is: I have a little doubt on the viability of RIPEMD160, by the curse of "birthday paradox", approaching 2^80 computation doesn't seem to be too remote. It's just not economical (profitable). At present, the whole network can do 2^66 hashes per second, it takes 2^14 seconds - less than 1/4 day, to reach 2^80 computation. And remember, that is 2 times SHA256 on a block. ASIC's that are computing one-time random SHA256 should cost less and do more.

finding a hash collision for RIPEMD160 is impossible because the chances of it are astronomically small and it  doesn't do you any good because you need to have a private key that produces a public key that is then hashed with SHA256 and then hashed with RIPEMD160 that produces the same hash! that is the combination of 3 impossible things!

2^80 is huge and what miners are doing with their 2^whatever hashes per second is not a hash collision find!

You don't need to go through the same steps of generating an address, since you know the public key space is 160 bit long, you just need to search 160 bit long private key space, in a hope to find a collision.

Example: If I know you are mapping a value in the space of one trillion, 2^40, to a space of 2^4 (0-F in hexdecimal), I just need to try all 2^4 numbers to see which one collides, I don't need to know or compute the original number in the range of 2^4.

Bitcoin is mapping 256 bit key space to 160 bit space, actually the public key space is 2^96 times smaller than its private key space, - much more shrinking of co-domain than the above example 2^40 to 2^4 (only 2^36 times smaller) 

I agree 2^80 is a huge number, actually the world's fastest computer can do 200K trillion flops, that's close to 2^58, to generate 2^80 keys, that machine will need 2^22 seconds, which is 46 days. Then you will have about 50% that there exist collided keys inside this huge list of 2^80 keys. I see 2^80 is the magic number that will put Bitcoin on the brink of collapsing, sure we are not quite there yet, but are we too far from it? Considering the rate the computers/GPU's/ASIC's are produced, I don't think the scenario is too remote. 

[quantummachine]

Database updated May 7, 2019

Address in total: 24,065,464 addresses that have Bitcoin balance.

[zephuros]

yo send me a dm when you can

[MrFreeDragon]

Here is the daily updated dump (address and balance): https://gz.blockchair.com/bitcoin/addresses/

[Zemomtum]

This is a brilliant effort and will be a good source for those that are doing research on Bitcoin and other alternative coins

[prospectorr]

This is a brilliant effort and will be a good source for those that are doing research on Bitcoin and other alternative coins

Who is going to benefit from this balance in wallets? does it make sense without the original private keys?

[stomachgrowls]

This is a brilliant effort and will be a good source for those that are doing research on Bitcoin and other alternative coins

Who is going to benefit from this balance in wallets? does it make sense without the original private keys?

No one does. No keys = No wallet.This is just a similar site of https://keys.lol/ and others where it listed out keys but dont actually had those balances.

It isnt bad to check out though but finding a collision or an address does have balance do have the slightest chances to happen.