Vanity Forge — Address Hunt & Hardware Benchmark
Hunt an Ethereum address starting with your word using every core on your machine — and find out how many keys per second it really does.
Hunt an Ethereum address that starts with your word — on your own hardware.
- 0/sSpeed
- 0Keys tried
- ∞Expected
- 1 in 66kDifficulty
- 8Cores
How to play
- Pick a hex word (0-9 and a-f only, so try beef, c0ffee, dad).
- Every core on your machine starts generating real keypairs and checking them. Nothing is uploaded — the keys are made and discarded in your browser.
- Each extra character makes it 16× harder. The counter is your machine’s honest key-generation speed, so this doubles as a hardware benchmark.
How it works
Pick a hex word and every core on your machine starts generating real Ethereum keypairs and checking whether the address begins with it. The reels lock as deeper matches are found, and the counter shows your machine’s honest key-generation rate, which makes this a hardware benchmark as much as a hunt. Each extra character multiplies the difficulty by 16, so four characters is seconds and seven is a long afternoon. Everything runs in your browser: keys are generated and discarded locally, and nothing is ever uploaded.
Keys shown on a screen must be considered burned — generate a fresh address in a real wallet before putting funds anywhere near it.
Frequently asked questions
Is the speed a real benchmark?
Yes. The counter is actual keypairs generated and checked per second by your own CPU, across every core the browser gives us minus one for the interface. It is a fair way to compare machines.
Why does one more character take so much longer?
Each character is one hex digit, so it multiplies the search space by 16. A four-character target averages 65,536 tries; seven averages over 268 million.
Is it safe? Where do the keys go?
Nowhere. Key generation happens entirely in your browser using audited libraries, and the discarded keys never leave the tab. Still, never fund an address whose private key has been displayed on a screen.