How long a pump.fun coin takes to graduate, and which deployers graduate in the first slot
Of 108,901 pump.fun graduates since July 22, 2026, 37.9% filled their bonding curve within two slots of being created. The rest took a median of 7 minutes. Which deployers graduate instantly, which never do, and what the first day of trading looks like for each.
Quick answer
From July 22 to October 31, 2026, 3,917,381 coins were launched on pump.fun and 108,901 of them graduated (2.78%). They graduate in one of three ways.
- The launcher alone, in the create's own slot: 26,539 graduates (24.4%). The deployer's wallet buys the whole curve, 85.01 SOL, in the transaction that creates the coin. One buyer.
- A handful of wallets in the first slot: 14,736 graduates (13.5%). The curve is filled in the create's slot, or the next two, by a median of 7 wallets, and the deployer is not the one who completes it.
- Climbed: 67,626 graduates (62.1%). The curve fills by trading, with a median of 37 buyers, in a median of 6.9 minutes: a quarter take under 2.6 minutes, a quarter over 38.9 minutes, and 9.0% more than a day.
So 37.9% of graduations happen before anyone outside the launch can see the coin. A deployer who launched one coin is the most likely to graduate this way (69.0% of their graduates); a deployer with 100 or more launches almost never does (1.7%).
Every day's graduations, with these timings and the deployer's record, are free at GET /graduations/{day}.json.
How we measured it
For every coin created since epoch 1006 (July 22) whose bonding curve completed, we take two events from the chain: the coin's create and its curve's complete, the moment the last token on the curve is bought. The gap between them, in slots and in seconds, is the time to graduate. pump.fun's migration to its pool follows the complete within seconds (median 0, 99% within 4), so the migration time on a coin page is a fair stand-in.
The shape is decided by the slot gap and by which wallet's buy completed the curve. Buyers are the distinct wallets that bought on the curve up to the complete.
The three ways to graduate
| how it graduated | graduates | share | median buyers before the complete | median SOL into the curve | median time |
|---|---|---|---|---|---|
| the launcher alone, in the create's slot | 26,539 | 24.4% | 1 | 85.01 | 0 s |
| a handful of wallets in the first slot | 14,736 | 13.5% | 7 | 85.01 | 0 s |
| climbed | 67,626 | 62.1% | 37 | 32.26 | 416 s |
The first row is the pattern we wrote up on October 24 in A quarter of pump.fun graduates are bought whole by their own launcher: one transaction creates the coin and buys all of its curve.
The second row is new. Of these 14,736, 13,064 were completed in the very slot the coin was created in and 1,672 in the next two slots. A median of 7 wallets bought before the curve was full; none of them was the deployer completing it. 5,115 different wallets made the completing buy, and they repeat across deployers: 91 wallets completed the curve for 10 or more different deployers, 8,993 coins between them. The busiest, 6A59APPsPXCBbMZJsAKfMSf1tiXnYFxWM4Ho3b78dAeP, completed 1,362 coins for 1,362 different deployers; the next two did it for 637 and 553.
The climbed curves took longer, and unevenly:
| time to graduate (climbed curves) | |
|---|---|
| a quarter graduated within | 2.6 min |
| median | 6.9 min |
| three quarters graduated within | 38.9 min |
| graduated within 10 minutes | 57.7% |
| graduated within an hour | 78.0% |
| took more than a day | 9.0% |
Which deployers graduate instantly
The deployer's own history decides the shape more than anything else. By how many coins the deployer launched in the window:
| deployer's launches | graduates | in the first two slots | by the launcher alone | median time when it climbed |
|---|---|---|---|---|
| 1 | 52,821 | 69.0% | 44.6% | 410 s |
| 2 to 10 | 13,905 | 26.6% | 17.0% | 694 s |
| 11 to 100 | 14,959 | 4.7% | 3.1% | 481 s |
| 100 or more | 27,216 | 1.7% | 0.5% | 326 s |
Instant graduation is a one-coin wallet's move. 24,676 different wallets were the launcher in a launcher-alone graduation, and 23,570 of them launched nothing else in the window: a fresh wallet per coin, as the October 24 report found.
Among the 865 deployers with 10 or more graduates, 27 graduated every one of them in the first two slots, and 767 never once did. The five with the most graduates all climbed:
| deployer | launches | graduates | in the first two slots | median time to graduate |
|---|---|---|---|---|
bwamJzztZsepfkteWRChggmXuiiCQvpLqPietdNfSXa | 31,643 | 608 | 1.3% | 97 s |
7FVfSdnR9VPGjMtmBP1Hz9C2DFTpoNX8gVVRmnimnGt9 | 8,692 | 440 | 0% | 422.5 s |
3QsuQtczCa7WQQK7LvFu2m8CDFazzwqxmcVDUts6ZMpP | 1,231 | 296 | 0% | 774 s |
CfBoFQ3tRrKhhjoXiocdVFJ9WkCzQNwbEK8uBZ6vRnrR | 1,235 | 217 | 0% | 206 s |
2E94st2NZnzA943HBceijgkw75gXTTxch39yquMBfeQk | 378 | 202 | 0% | 443.5 s |
And the deployers who graduate everything in the first slot:
| deployer | launches | graduates | in the first two slots |
|---|---|---|---|
H6zpaY14WgkEmB3MJTvC7Wb5PQJkBVLd4KwLD2ktVgop | 40 | 39 | 100% |
D8pP1hegfQrND1tpMv1pQJZYyzwnHsiFa9AVLc2Lzoxy | 31 | 31 | 100% |
3AF3YaZf1g5ycPim3QtWFYiELc8mh7SvRLK83P5xDjNn | 24 | 24 | 100% |
3sgR3SeQC4JxSDZFdaTY5ezDYxjZ8rp1BCncgFy5dmzB | 24 | 24 | 100% |
DrZhh8Uz9ahWYEXyv1zWNgemN3BRAi42aSZVV2vsKsVy | 23 | 23 | 100% |
The serial deployers, the ones with hundreds or thousands of launches, graduate few coins and climb for every one. The deployers who graduate instantly are either fresh wallets or a few dozen wallets that do nothing else.
The same names, hundreds of times
The coins graduated in the first two slots reuse names. 63 tickers were each used on 100 or more of them, 14,570 coins between them. The most used:
| ticker | coins graduated in the first two slots |
|---|---|
| FOMO | 705 |
| WOFI | 648 |
| TNOS | 563 |
| USWR | 513 |
| NTDA | 481 |
| GTA 6 Coin | 448 |
| MrBeast | 443 |
| USOH | 392 |
A ticker on a coin page says nothing about which of its hundreds of copies you are looking at. The copycat check lists the others by address.
The first day of trading
For each SOL-quoted graduate, the trades in its own PumpSwap pool in the 24 hours from its create:
| how it graduated | coins | median trades | median wallets trading | median trade | trades under 0.01 SOL |
|---|---|---|---|---|---|
| the launcher alone | 25,730 | 10,291 | 2,112 | 0.0004 SOL | 75.8% |
| a handful of wallets in the first slot | 14,707 | 4,357 | 938 | 0.1366 SOL | 12.5% |
| climbed | 56,349 | 666 | 148 | 0.058 SOL | 21.0% |
The launcher-alone coins have the busiest first day by count: a median of 10,291 trades from 2,112 wallets, fifteen times a climbed coin's. But the median trade on them is 0.0004 SOL, and three trades in four are under 0.01 SOL. A trade count or a holder count on these coins measures the number of very small trades, not the number of people.
The first-slot coins are different: fewer trades than the launcher-alone ones, but a median trade of 0.1366 SOL, larger than on coins that climbed. What a buyer an hour in can expect from launcher-alone coins is in the October 24 report.
Month by month
| month | graduates | the launcher alone | first slot, a handful of wallets | median time when it climbed |
|---|---|---|---|---|
| July (from the 22nd) | 8,585 | 2.0% | 25.1% | 492 s |
| August | 33,622 | 19.6% | 12.9% | 405 s |
| September | 34,093 | 31.9% | 13.5% | 432 s |
| October | 32,601 | 27.4% | 11.1% | 387 s |
The launcher-alone share rose through August and peaked in September; the first-slot share was highest in late July and has held near an eighth since. A climbing curve takes about as long as it did in July.
Frequently asked questions
How long does it take a pump.fun coin to graduate?
If it climbs, a median of 6.9 minutes from its create: a quarter graduate within 2.6 minutes, three quarters within 38.9 minutes, and 9.0% take more than a day. But 37.9% of graduates do not climb at all: their curve is full within two slots of the create.
What does it mean when a coin graduated in the same second it was created?
Its whole bonding curve, 85.01 SOL, was bought before anyone outside the launch could see the coin: by the deployer alone in 26,539 cases, by a median of 7 wallets in 14,736. Either way the supply sits with very few wallets from the first second.
Which deployers graduate instantly?
Mostly wallets that launched one coin: 69.0% of their graduates filled in the first two slots. Deployers with 100 or more launches did it on 1.7% of theirs. Among the 865 deployers with 10 or more graduates, 27 did it every time and 767 never did.
How do I see this for a coin or a day?
GET https://pumpdata.fun/graduations/{YYYY-MM-DD}.json lists every graduate of a UTC day (the last three days, free, no key) with graduated_after_s, same_second, the deployer's launches and graduations before it, and the day's top graduating deployers. For one coin, the check page shows the deployer's record, and created against migrated_at in GET https://pumpdata.fun/free/launch/{mint} gives the gap.
Every number here comes straight from the blockchain, the same data behind the paid calls: the calls ($0.005 to $0.01 each, USDC over x402, no account), yesterday's /stats and the daily wire are free.