Pump.fun changed its PUMP buyback policy on April 29, 2026. The old setup routed 100% of bonding curve, PumpSwap, and Terminal revenue into open-market PUMP buys that were then burned. The new policy splits revenue 50/50: half flows into an irreversible smart contract that buys and burns PUMP forever, half stays with the company. By the time of the switch, roughly 36% of PUMP supply had already been burned, wiping out about $370 million in market value.
The Short Version
The 100% buyback model didn’t hold the PUMP price up the way the team wanted. The token kept sliding even as the burn rate accelerated. The April policy update cut the buyback rate in half but locked the remaining 50% into a smart contract that can’t be changed, retired, or paused. The other 50% pays for product work, hiring, marketing, and acquisitions. The bet is that a smaller but credible burn signal, plus a funded company, beats a bigger burn that the market never believed in.
What the Old Policy Looked Like
When Pump.fun launched PUMP in mid-2025, the pitch was straightforward: every dollar of platform revenue would buy PUMP on the open market and burn it. Revenue came from three places:
- Bonding curve fees — 1% of each on-curve trade
- PumpSwap swap fees — 0.05% of each post-graduation swap (out of the 0.25% total)
- Terminal revenue — Subscription and premium feature revenue from the trading terminal
All of it bought PUMP at market and sent it to a burn address. The burn ran continuously, in small chunks, all day. The team reported burn numbers weekly on socials and they were genuinely large. About 36% of total supply was burned by late April 2026.
The policy failed by its own metric. PUMP’s price slid through most of Q1 2026 despite the relentless burn. Annualized 2026 revenue was tracking around $320 million, down from $971 million in 2025. A shrinking revenue pool burning a shrinking dollar share of supply was producing a smaller per-month deflation rate than the chart implied at first glance.
What Changed in April 2026
On April 29, 2026, Pump.fun published the new policy:
- 50% of all net revenue flows into an irreversible smart contract that buys PUMP on the open market and burns the tokens. Nobody, including the Pump.fun team, can change this allocation.
- 50% of net revenue stays with the company. It funds product development, payroll, marketing, runway, and potential M&A.
The buyback contract is the key piece. The old burn was a company policy that the company could change at any moment, which it did. The new buyback is a smart contract with no admin key. Pump.fun cannot pause it, redirect it, or cut its share. That’s what “irreversible” means in this context.
The mechanics inside the contract are unchanged from the old burn. Revenue flows in, the contract submits market buys for PUMP, the bought tokens go to a burn address. The frequency, sizing, and slippage logic are similar to the old setup. What’s different is the share of revenue feeding it and the lack of an off switch.
Supply Math After the Change
A few numbers worth pinning down:
| Metric | Value |
|---|---|
| Original PUMP supply | 1,000,000,000,000 (1T) |
| Burned by April 2026 | ~360B (36% of supply) |
| Effective circulating supply post-burn | ~640B (varies with vesting) |
| Burn dollar value before policy change | ~$370M |
| Revenue annualized 2026 | ~$320M |
| Share of revenue now buying PUMP | 50% |
At a 50% revenue share and roughly $320M annualized revenue, the buyback contract has on the order of $160M per year to spend on PUMP. The actual amount fluctuates with platform volume, SOL price, and fee mix. The contract converts revenue into PUMP burns at whatever the market price is on the day.
The exact deflation rate matters less than the floor it implies. Even if PUMP price falls, dollar revenue keeps buying tokens at the new lower price, which removes more units per dollar. That self-correcting effect is part of why the team kept the burn structure intact instead of switching to a fixed token amount per period.
Why Half Instead of All
The team’s argument was that the 100% policy left the company underfunded. Pump.fun was profitable on paper but couldn’t reinvest in product or team. Hiring slowed, marketing shrunk, and rivals (LetsBonk, LaunchLab, smaller pads) shipped faster while Pump.fun stayed cash-poor on the operating side.
The 50% split was framed as a sustainable balance. Half goes to holders through deflation. Half keeps the platform competitive. The view from the team was that a slower but reliable burn from a healthier company is worth more long-term than a faster burn from a starving one.
The market response was mixed. PUMP dipped on the announcement, recovered some of the loss within a few days, and has traded in a tighter range since. The irreversibility of the contract was the part that did the most work in defending the price. A revocable 100% policy is worth less than a permanent 50% one.
Where Revenue Comes From in 2026
The buyback feeds off the full Pump.fun revenue stack:
- SOL bonding curves — 1% on-curve fee, still the largest revenue line
- USDC bonding curves — Added May 21, 2026. Same 1% on-curve fee, see the Pump.fun USDC pairs guide
- PumpSwap fees — 0.05% protocol share on every post-graduation swap
- PumpSwap LP migrations — Now free at graduation, so no migration revenue
- Terminal subscriptions — Premium features, alerts, and pro-tier access
USDC pools widened the revenue base without changing the split. Every USDC dollar of fee revenue flows through the same 50/50 allocation. The team confirmed this explicitly when launching the USDC pair option.
How the Buyback Compares to Other Token Burns
Several Solana ecosystem tokens run buyback programs. Quick context:
| Project | Revenue source | Buyback share | Reversible? |
|---|---|---|---|
| Pump.fun (PUMP) | Curve fees, swaps, terminal | 50% | No (smart contract) |
| LetsBonk (BONK) | Curve fees, trader rewards | ~50% to BONK buybacks | Policy, reversible |
| Jupiter (JUP) | Aggregator fees | Varies, governance-driven | Policy, reversible |
| Raydium (RAY) | Swap fees | ~12% buyback share | Policy, adjustable |
PUMP is the only one with an irreversible smart contract enforcing the share. The others are policies the teams can change. That’s the strongest single argument for PUMP as a deflationary asset compared to peers: the contract takes the decision out of human hands.
What This Means for Traders
For a meme coin trader, the buyback doesn’t change your trading day. The bonding curve, PumpSwap fees, and bot behavior all work the same way. The buyback runs in the background. It matters mostly if you hold PUMP itself or if you trade tokens whose price moves with sentiment on Pump.fun’s longevity.
Three practical takeaways:
- The platform is funded. Revenue retention solves the underfunded-company problem. Expect more product work over the next year, not less.
- The burn is mechanical. Buy decisions follow revenue, not discretion. A high-volume month burns more. A low-volume month burns less.
- PUMP price still mostly tracks meme coin volume. The buyback is a tailwind, not a pump. PUMP rallies require meme season volume more than they require burn announcements.
Risks and Open Questions
Several things could still derail the model:
- Volume decline — If Pump.fun loses share to LetsBonk or LaunchLab, the revenue base shrinks and the burn shrinks with it. The contract is permanent. The dollar flow into it is not.
- Regulatory risk — Pump.fun faces a pending lawsuit in the US around its early launch mechanics. Adverse rulings could affect operations and revenue.
- Token unlocks — Original team and investor allocations vest over time. New supply hitting the market can offset burn pressure on price.
- Competitor responses — LetsBonk’s 50% BONK buyback is already pulling some creators. If LetsBonk widens the rewards model, share could shift further.
The 50/50 split shipped at a moment when Pump.fun’s competitive position was the most contested it has ever been. The policy will be tested over the next several quarters by volume, competition, and regulation.
FAQ
Can Pump.fun change the buyback share back to 100%?
It can change the company-side share, but not the contract-side share. The 50% feeding the buyback contract is locked. The other 50% is company-controlled and could be redirected.
Is PUMP deflationary now?
The buyback burns PUMP every day, so the circulating supply trends down over time as long as revenue exists. Vesting unlocks push the other way. Net effect depends on the balance.
How much PUMP gets burned per day?
It varies with revenue and PUMP price. Roughly, half of daily fee revenue buys PUMP at market and burns it. At current revenue and price levels, that’s in the low single-digit millions of dollars per day.
Does USDC volume burn PUMP?
Yes. Revenue from USDC pools flows through the same 50/50 split. The buyback contract uses the USDC revenue share to buy PUMP on the open market and burn it.
Where do I see the burn live?
The buyback contract address is public, and on-chain burns are visible in real time on Solana block explorers. Dashboards on dexscreener.com and dune.com track cumulative burn totals.
Is PUMP a good buy because of the burn?
The burn is one input among many. Volume trends, competition, vesting unlocks, and regulation matter at least as much. The burn is structural support, not a guarantee of price appreciation.
What to Read Next
- Pump.fun USDC Pairs Guide — How the new USDC curves feed the buyback
- Pump.fun vs LaunchLab vs PumpSwap — Where Pump.fun sits in the Solana launchpad stack
- PumpSwap Trading Guide — Trading the destination DEX
- LetsBonk vs Pump.fun — The main competitor with its own buyback model
- Building Your Solana Meme Trading Stack — Tool setup for the current launchpad mix
Disclaimer: PUMP token economics can change with policy updates, governance decisions, or regulatory action. This guide is educational, not financial advice. Always verify current contract details and revenue numbers from primary sources before trading. See our full Risk Disclaimer.
