buyback model still the most effective this cycle
Every effective token with buyback mechanism at the moment.
Buybacks are one of the strongest token mechanisms in the current market.
But most buyback lists include programs that are small, paused or still waiting to launch.
I have been tracking the actual fee-to-token flow instead.
My criteria are simple:
– Live now
– Funded by recurring protocol fees
– Programmatic or verifiable onchain
– Large enough relative to circulating supply
Based on industry trackers, protocols have spent around $638M on buybacks in 2026 YTD.
@HyperliquidX and
@Pumpfun account for roughly 90% of that amount.
These are the five tokens I currently consider part of the main buyback meta:
[1]
$HYPE
Hyperliquid routes around 97-99% of trading fees to the Assistance Fund.
Cumulative purchases are estimated at $1.1-1.3B, including around $370M in 2026.
AQAv2 will also allocate roughly 90% of USDC reserve yield, with the first payout expected on October 3.
$HYPE remains the largest fee-funded token buyer by dollar value.
I consider it the strongest model because the buyback is supported by a large and recurring trading business.
The main variable is emissions. A large buyback does not guarantee net deflation if unlocks remain higher.
[2]
$PUMP
Pumpfun currently allocates 50% of net revenue to buybacks, down from 100% before April 28.
Purchased tokens are burned immediately.
Around $400-445M has been spent cumulatively, removing approximately 16% of the maximum supply.
Estimated 2026 spending is already around $200M.
I like the transparency of the burn, but
$PUMP also proves that buybacks cannot support price alone.
Revenue durability, valuation and remaining unlocks still matter.
[3]
$PONS
@ponsdotfamily routes 80% of its protocol fee share into buying and burning
$PONS.
Around 29-30% of the 1B supply has already been burned.
Its latest 30-day fees were approximately $119M, with around $11.6M flowing toward holders and buybacks.
I find
$PONS particularly interesting on percentage terms.
It operates at a much smaller valuation than HYPE or PUMP, but it has already removed a larger share of supply than most launchpad tokens.
The question is whether
@RobinhoodCrypto launch activity can remain high after the current attention cycle.
[4]
$ASTER
@Aster_DEX allocates 99% of daily platform fees to purchasing
$ASTER.
The purchased tokens go to veASTER stakers, while an equal amount is burned from protocol reserves.
The team describes this as a 198% allocation model.
Listing fees also fund additional purchases.
This is one of the most aggressive designs I have seen from a perp DEX.
I am watching whether Aster can maintain enough trading volume to make the percentage meaningful in dollar terms.
[5]
$STONK
@LaunchOnSF routes roughly 60% of platform revenue into buying and burning
$STONK.
Despite launching only in August, it has generated more than $9M in lifetime revenue and burned around 14% of supply.
It also briefly exceeded Pons in daily revenue.
This is still a young protocol, so I need more data before treating the current revenue level as sustainable.
Outside these 5, several serious programs remain active:
–
$AAVE: approximately $1M purchased weekly and held
–
$UNI: protocol fees fund buy-and-burn
–
$RAY: long-running buy-and-burn that has historically offset emissions
–
$SKY: Smart Burn Engine, with 5% of monthly net surplus now allocated to buy-and-burn
–
$JUP: 50% of fees used for purchases held for three years
–
$JTO: 100% of Jito Network revenue allocated through its Cryptoeconomics SubDAO
– gmx:native, ethereum:0xc20059e0317de91738d13af027dfc4a50781b066,
$MET, ethereum:0x046eee2cc3188071c02bfc1745a6b17c656e3f3d, dydx-chain:native and
$ETHFI also operate active programs
I do not put all of them in the same category.
Some burn tokens. Some hold them in a treasury. Some distribute them to stakers. Each structure creates a different effect on liquid supply and holder value.
My framework has five checks:
– How durable are the protocol fees?
– What percentage of revenue reaches the token?
– Are purchased tokens burned, locked, held or distributed?
– Do buybacks exceed unlocks and emissions?
– Is the buyback large relative to market cap and circulating float?
That last comparison matters most.
Tokenomist’s mid-2026 analysis found that only
$BNB and
$RAY were clearly reducing net supply after emissions.
$HYPE can buy hundreds of millions of dollars while unlocks still expand supply.
$PUMP can burn 16% of its maximum supply while trading far below ATH.
Buybacks improve token economics. They do not replace revenue growth or reasonable valuation.
My current buyback watchlist is:
$HYPE,
$PUMP,
$PONS,
$ASTER and
$STONK.
i am most interested in protocols where recurring fees, a high allocation rate and controlled emissions operate at the same time.