15 of 128 Token Addresses End in 1e18

While running today's name-collision survey on Robinhood Chain, I also counted the last four characters of every token contract address I pulled. The result was not what I expected: out of 128 sampled tokens, 15 addresses end in 1e18 and 5 end in 7777, while every other four-character ending appears at most twice.

That is not chance. Somebody is mining these addresses on purpose. Here is what the cohort is, what it does and does not prove, and whether a vanity address is worth paying for when you launch.

The list

Data pulled 2026-09-18, 00:20–00:50 UTC, from DexScreener (liquidity and volume are totals across every pool that site indexes for the token).

Two tokens from today's hot list are in that table: AI (Artificial Inu, +25.44% today) and MEME (A Meme Coin, +23.74%).

Why 1e18

Anyone who has written a contract gets it instantly: 1e18 is ten to the eighteenth, the constant behind 1 ETH = 10^18 wei. It shows up in nearly every ERC-20 (the standard 18 decimals). As an address ending it is a signature aimed at people who read code.

The 7777 group is the plainer kind of lucky number: mooncat 0x862c…447777 ($216,416), ZEC 0x725d…757777 ($25,781), Tokenization 0x6162…F77777 ($9,430), NFT 0x8068…C37777 ($9,389), SHINYPACK 0x555A…367777 ($9,383).

How the addresses get mined

In an ordinary deployment, the address is derived from the deployer address and a nonce, so you do not get to pick it. To pick it you use CREATE2: address = deployer + a salt you choose + the hash of the contract bytecode. Change the salt and the address changes, so you can grind salts locally until one produces an ending you like, then deploy with that salt.

The cost is compute. Each extra hex character multiplies the average work by 16. A four-character ending like 1e18 takes roughly 65,536 attempts on average — seconds on a laptop. Six characters, like that …F77777, takes real time. Four-character vanity is cheap, which is exactly why fifteen of them show up at once.

What this proves and what it does not

Worth drawing the line carefully, because this kind of statistic invites over-reading.

What it supports:

• In these 128 samples, 1e18 appears 15 times and 7777 five times, while no other four-character ending appears more than twice. That distribution is not random — by chance, any specific four-character ending would be expected to appear under 0.02 times in 128 addresses.
• So these addresses were almost certainly mined with CREATE2 rather than landing there by accident.

What it does not support:

• It does not show they came from the same people. Address grinding is something anyone can do with public tooling. A shared ending indicates shared taste or shared tools, not shared ownership.
• It does not give a chain-wide rate. These 128 tokens came from twenty keyword queries against DexScreener's search endpoint — a convenience sample, not a census. It can show the ending is wildly over-represented; it cannot tell you what percentage of the chain looks like this.
• It does not mean vanity equals legitimate. Look at the right-hand column: six of the fifteen did under $1,100 of 24h volume, including "To The Moon" at $2, "token" at $8 and "I Like The Coin" at $10. The effort spent picking an address has nothing to do with whether anyone buys.

So what is a vanity address actually good for

For a launcher, it bundles three small things:

1. Recognisability. This chain is crawling with name collisions right now — seven Zcash contracts, four Shiny Packs, two Artificial Doges, all found today. A memorable ending lets your pinned post say "ends in …bbb", which is far more realistic than asking people to read 42 characters.
2. A credibility signal. Paying a little to pick an address suggests you did not spin this up in three minutes. It is a signal, not a guarantee — the table above proves that.
3. Copy resistance. Anyone can clone your name for free. Nobody can clone your address. The address is the one thing on-chain that cannot collide.

Vanity on BigPump: a bbb ending for 0.001 ETH

BigPump is a Tax RWA/NFT bonding-curve launchpad on Robinhood Chain, and its launch flow has an optional box: make the token contract address end in bbb for 0.001 ETH. The mechanism is the CREATE2 salt grinding described above, just folded into the launch so you do not run the script yourself.

While we are on launch mechanics, the rest of the numbers, for comparison (these are BigPump's own parameters):

• One-click launch, no code, and no initial liquidity required from you
• Bonding curve stage: a 1.5 ETH virtual reserve only sets the opening price; once cumulative real ETH reaches 4 ETH the token graduates automatically into a Uniswap V2 pool, and the LP is sent to the burn address
• During the curve there is only a 1% pool fee (95% platform / 5% creator). The creator tax (1/3/5/10%, split four ways across treasury / burn / dividends / liquidity) and holder dividends only apply to DEX trades after graduation
• Graduation takes a one-off 5% of the ETH in the pool; after graduation the platform takes 1% per DEX trade
• Dividends are paid in ETH pro-rata to holdings above a minimum threshold; pair, router, burn and protocol contracts are excluded
• The token is an ERC-20 (6 decimals) bound to an ERC-1155 NFT, where 1 NFT = 1,000,000 tokens

Details at book.bigpump.ai; the launch page is /robinhoodchain/create.

For buyers: how to use an address ending

Treat it as a verification tool, never a filter.

• ✅ The project's pinned post says the contract ends in …1e18, and you glance at the ending before signing. Faster than reading 42 characters, and it catches most lazy same-name copies.
• ❌ "The address is a vanity address, so the project is solid." Today's $2-of-volume "To The Moon" is the counterexample.
• ⚠️ Endings can be copied too. Someone else can grind their own …1e18 address to ride yours. Eventually you still have to match the full string.

FAQ

What does 1e18 mean?

Ten to the eighteenth — the constant behind 1 ETH = 10^18 wei, and the origin of ERC-20's standard 18 decimals. As an address ending it is an in-joke for developers.

Are the fifteen 1e18 tokens from one team?

There is no public data establishing that. Address-grinding tools are public, and a shared ending is not evidence of common origin. I am not going to speculate.

What does mining a vanity address cost?

A four-character ending averages about 65,536 attempts — seconds on a normal machine. Each extra hex character multiplies the average work by 16, so four is cheap and six starts to get expensive.

Does a vanity address mean the project will not rug?

No, not at all. Six of today's fifteen did under $1,100 of 24h volume. A pretty address and an actual market are unrelated facts.

Where do I check these live?

Opening a trading page by contract address is the safest route — for example the $AI trading page. Today's full board is in the September 18 hot tokens report, and the method for telling copies apart is in Two Artificial Doges.

BigPump is an independent project, not affiliated with, endorsed by or sponsored by Robinhood Markets, Inc. The token named "Robinhood" in the table above is a third-party deployment on-chain and is likewise unrelated to Robinhood Markets, Inc. On-chain figures come from a DexScreener sample taken 2026-09-18, 00:20–00:50 UTC. Not investment advice — memecoins can go to zero.

Disclaimer: memecoins are extremely volatile and most go to zero. This article is not financial advice. Do your own research and only spend what you can afford to lose.