Solana · fee-funded distribution
Every trade pays a fee. The fees buy $ANSEM on the open market. The $ANSEM goes to holders, weighted by how long you actually held — not by who happened to be holding at one lucky instant.
not launched
Claim
Your $ANSEM accumulates from the first epoch you held through. Claim whenever you like — today, or in six months. There is no per-epoch button and nothing expires.
You claim your own, and you pay for your own: roughly four hundredths of a cent in gas. Nobody is ever asked to fund a payout for anyone else.
Distribute fees is the other half, and anyone may press it — holder or not. It sweeps trading fees out of pump into the vault, and pays whoever runs it nothing: the guarantee is that nobody can stop you, not that you are paid. We forward the vault into the treasury every 12 hours, which is the one step that needs us.
The mechanism
Status: nothing has launched. No token, no pool, no epoch yet. This documents a mechanism that has been built and tested — not one that is running.
What we did instead
Every epoch credits every holder their exact share, no minimum. Credits accumulate. A wallet enters a distribution once its own balance beats its own claim cost. Nobody is dropped.
Balance integrated across the whole epoch, sampled at randomized times never published in advance. Buying before a snapshot and selling after earns approximately nothing.
What you trade against and what you buy with fees are separate decisions. Collapse them and you end up quoted in something no chart or router indexes — then earn nothing.
Pump's fee collection takes no signer at all — anyone can trigger it, and the proceeds are pinned to the creator vault by address derivation. Not a promise we make; a property of the program.
Fees land in a Squads vault, and moving them to the treasury needs our signature. That hop runs every 12 hours after launch. It is deliberate: the flywheel has no withdrawal instruction, so paying fees straight in would strand them forever if the unaudited code has a defect. Recoverable beats trustless until the audit says otherwise.
Proof
Each epoch publishes snapshot slots, per-wallet time-weighted balance, the exclusion list, the swap and the merkle root — enough to rebuild the tree and check your own allocation. The tree structure follows the audited Jito merkle distributor, checked against its Rust vectors — but is deliberately not wire-compatible: our leaf also commits the recipient's index.
Rebuild it yourself → Recompute the whole tree in your browser and check any wallet's proof.
Testing
collect_creator_fee and collect_coin_creator_fee take no signer — read from the published IDLs. The destination is fixed by PDA derivation, so anyone can sweep fees and nobody can redirect them. The one step this does not cover is moving them from the vault into the treasury, which needs us and runs every 12 hours.25vupHr3BEW2qALudZiic1RVNDTNBhxBf7QSFLjc63n8 — live on mainnet, baked in at mint, unchangeable after. Publishing it now is a commitment we cannot quietly walk back.Not yet done: nothing deployed to mainnet, and no formal audit — audit status →
Audit
Five instructions and three accounts — every path that touches money, including the swap, the published root and the claim. The surface is deliberately small so there is less of it to get wrong: no per-epoch accounts, no clawback, and no instruction by which anyone, the admin included, can move treasury funds.
Findings were returned, fixed and re-tested before this page went up. The specifics stay with the auditors until a formal audit is complete.
Every fix is covered by 34 checks running against a live validator, alongside the unit tests — not by reading the change and calling it done.
Then the tests were tested. Each new guard was deliberately removed and the suite re-run, to confirm it went red. A guard no test can catch failing is not a guard.
Reviewed with CertiK’s AI audit tool. That is not a CertiK certification, and not a formal CertiK audit — both are still pending. An AI review is a floor, not a ceiling: it reads the code it is shown and cannot tell you what a live market will do to it. Until a formal audit is complete, treat this as unaudited software holding real money.
Do your own research. None of this is financial advice. Tokens are risky assets and this one is no exception — it can go to zero, distributions depend entirely on trading fees that may never materialise, and smart contracts can fail in ways no review anticipates. Never commit more than you are willing to lose completely.
Roadmap
A date is a promise about work nobody has done yet. This is the order things happen in, and where we actually are.
Phase 04 is conditional on the token earning fees at all. Our own research says most launches earn almost nothing — that applies here too, and no roadmap changes it.
The Black Bull
ANSEMFUND is not live. Nothing here is an offer, a solicitation, or a promise of future distributions. Figures describing launch economics are measurements of other tokens' historical on-chain data, not projections for this one — the same data says the overwhelming majority of launches earn almost nothing, and that applies here too. Distributions depend entirely on trading fees that may never materialise.
The launch data → We sampled tens of thousands of pump.fun curves before building this, and read every Meteora pool ever created for the lifetime-fee picture pump does not record on-chain. It is about the odds any token faces, not about this one. Verify an epoch → Whitepaper →
Terms · Privacy · Disclaimer & risk
Unaffiliated. ANSEMFUND has no relationship with Ansem, who has publicly
disavowed tokens using his name, and none with pump.fun, Meteora, Jito or Jupiter beyond
using their public software and public on-chain data. $ANSEM refers to the token at
9cRCn9rGT8V2imeM2BaKs13yhMEais3ruM3rPvTGpump; several unrelated tokens share
that ticker. Not financial advice. Assume you can lose everything.