Pump.fun Launch Collector
Under maintenancePricing
from $2.00 / 1,000 results
Pump.fun Launch Collector
Under maintenanceEvery new pump.fun coin, priced from the second it's created. Reads the bonding curve on-chain and samples on a timer, so coins that die stay in your data.
Pricing
from $2.00 / 1,000 results
Rating
0.0
(0)
Developer
JJ Stowell
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Categories
Share
Every new pump.fun coin, priced from the second it is created.
Around 20,000 coins launch on pump.fun every day. This watches them being born and records what happens next — market cap, how much SOL people put in, and how much the creator bought themselves — sampled every few seconds from the moment of creation.
What you get
| Column | What it means |
|---|---|
symbol / name | The coin |
ageSeconds | How old it was at this reading |
marketCapUsd | Market cap in dollars |
changeFromCreationPct | Move since the launch price |
solInCurve | SOL that has actually gone in |
devBuySol | What the creator bought themselves |
devSharePct | The creator's share of all money in |
graduated | Whether it has left the bonding curve |
mint / pumpFunUrl | Address, and a link to the coin |
A typical row:
| symbol | age | market cap | change | SOL in | dev bought | dev share |
|---|---|---|---|---|---|---|
| Trio | 15s | $6,240 | +247.7% | 29.26 | 1.778 | 6% |
| TISUKI | 16s | $2,010 | +0.4% | 48.43 | 48.285 | 100% |
Those two rows are the reason people want this. Trio pulled in 29 SOL from real buyers in fifteen seconds. TISUKI's 48 SOL was all the creator's own money — the dev share column says so instantly.
Two things that make this different
Prices come off the chain, not from an API. pump.fun's own price field is wrong by tens of thousands of times on a minority of coins — one reported a $124M market cap on a coin its own API valued at $2,253. This reads the bonding curve account directly and matched the creation event to within 0.0000% when tested.
Sampling runs on a timer, not on trades. Feeds built from trade history have a hole shaped exactly like the answer: a coin that dies stops trading, so it stops producing rows and quietly disappears. Here every tracked coin is priced on a fixed clock whether anyone trades it or not, so the dead ones stay in your data.
That second point matters more than it sounds. A dataset assembled from trades tells you about the survivors. This one tells you what actually happened.
Settings
| Setting | Default | Notes |
|---|---|---|
| How long to run | 10 min | Longer catches more launches |
| Follow each coin for | 20 min | Each coin is priced for this long after creation |
| Seconds between readings | 10 | Lower is finer and costs more |
| Only coins with real activity | on | Skips the ~70% that never trade at all |
| Minimum SOL in the curve | 1 | What counts as "real activity" |
| Max coins at once | 200 | Ceiling so a launch storm can't run away with credits |
What people use it for
- Research — how launches behave in their first minutes, and how often they die
- Screening — spotting coins pulling real money in, versus a creator funding their own chart
- Backtesting — a clean history that includes the failures, which most feeds silently drop
Honest notes
About 70% of launches never receive a single trade. The filter is on by default so you are not paying for rows about coins nobody touched — turn it off if the death rate is what you are studying.
This is a data tool. It reports what happened; it does not predict anything and makes no claim that trading on it is profitable.
